AI-assisted delivery model

A small team.
A clear path to delivery.

I bring my enterprise delivery and leadership practice into the private MIN–MAX build: define the task, assign useful responsibilities, review the evidence and make the next decision.

From a product question to a release decision.

I set the outcome, route work through building and independent review, and decide the next authorized step. This is the delivery model for my private prototype.

01
Frame
I set the outcome, constraints and acceptance criteria.
Product brief
A bounded question and a testable brief.
Inspect the brief
02
Build
The assigned builder turns the brief into a candidate change.
Candidate change
The change, its assumptions and supporting checks.
Inspect the change
03
Review
Independent review checks behavior, evidence and readiness.
Review findings
Findings tied to the candidate being assessed.
Inspect the findings
04
Decide
I decide whether the evidence supports the next authorized step.
Approval decision
Approve, revise or hold, with a reason.
Inspect the decision

When a check fails: identify whether the problem is in the product, test, evidence process or environment, then correct and review that part.

01 · What I put in the brief

I define what should change for the user, the constraints the work must preserve and the evidence that would support the next decision. Scope and authority travel with the brief.

02 · What the candidate change includes

The candidate change comes with its assumptions, checks and unresolved questions. That lets the reviewer inspect the work against the same brief.

03 · What the review findings explain

Independent review connects expected behavior to observed results and identifies the layer that failed. Agent simulations check product paths; research with people answers questions about understanding and usefulness.

04 · How I make the approval decision

I retain approval for merges, deployments, beta access, production releases and expanded claims. A completed build or a passing check supplies evidence for that decision.

Explore roles and relationships Read the complete outline

Explore the full interactive map

Recorded MIN–MAX incident · Private prototype

A renter’s next step had to work.

I’m building MIN–MAX to help renters make apartment decisions. Independent review found a gap between the next step the interface offered and what the product could do. This is how I connect that problem to a correction and a release decision.

The product promise
The next action led nowhere.
“Review work-from-home requirement” appeared when the environment could not run that assessment. I needed the next step to resolve the renter’s question.
Inspect the failed action
My decision
The change had to stop.
I required the displayed action to match what that version could execute. The response was to stop the merge and correct the path.
Inspect my decision responsibilities
The correction
Connect the action to a working capability.
I used an action-availability contract: the assessment appears only when that version can run it. The defect was corrected in protected source.
Read the recorded correction
The next evidence
Verify the corrected path in deployment.
I still needed to verify that corrected path on the exact deployment. That requirement remained open at this stage.
Inspect deployment readiness

I apply the same question in production assignment: can the work reach the right eligible reviewer under priority and capacity constraints?

See the assignment correction in production

Inspect the full model: roles, filters and failure types

Interactive evidence map · executive progressive-disclosure view

One mission. Bounded roles. Independent evidence.

The Full Map connects 91 documented roles, workstreams, gates, incidents, lessons and outcomes through 174 evidence-labelled relationships.

View density

Search nodesTeamReadinessEvidenceTime horizon

Relationship layers

Critical path onlyShow handoffsShow blockersShow contingenciesShow my decision responsibilitiesShow customer impact

Accessible outline

Understand failure types and programme codes

Failure classification · plain language first

How we classify a failure

Identify the failing layer before choosing a fix. A product, test, evidence harness, integration pipeline and deployed environment answer different questions.

Product defect

The product behaves incorrectly, contradicts its intended decision model, creates a dead end, or presents an unsafe or misleading customer state.

Example: A page displays “Review work-from-home requirement” even though the environment cannot execute the work-from-home question.
Customer consequence: The visitor reaches a dead end or receives an untrustworthy decision.
Remediation boundary: May require product remediation.

Test defect

The test expectation, fixture, locator, or assertion is incorrect even though the intended product behavior works.

Example: A diagnostic expected the message field’s accessible name to equal one exact string even though the field was correctly labelled and described.
Customer consequence: Delivery is blocked by faulty evidence rather than faulty product behavior.
Remediation boundary: May require test remediation.

Evidence-harness defect

The mechanism used to create or retain evidence fails before it can validly judge the product.

Example: An evidence package could not start because its output directory or attestation-output contract was invalid.
Customer consequence: The product remains unadjudicated; no pass or failure may be inferred.
Remediation boundary: Requires valid evidence before judging the product.

Continuous-integration defect

The automated integration pipeline, test runner, or build orchestration changes or destabilizes the evidence environment.

Example: Automatic Continuous Integration (CI) Git-diff capture changed the checkout to a shallow repository after browser tests, invalidating later source-custody checks.
Customer consequence: Release evidence cannot be trusted until the pipeline behavior is corrected.
Remediation boundary: Requires pipeline or test-runtime remediation.

Environment defect

The deployed or test environment is incompatible with the approved source, schema, runtime, access, viewport, or binding contract.

Example: A QA database lacked required foundational tables, so the customer journey could not be executed end to end.
Customer consequence: Product behavior remains unproven in that environment.
Remediation boundary: Requires environment reconciliation or a new environment.

A later green run does not erase retained failed evidence. It shows that the corrected candidate passed its own exact gate.

Programme glossary

QA, CI and agent-simulation codes

Quality Assurance (QA)QA: QA means Quality Assurance. The independent validation activity or role that challenges source and intended product behavior. Quality Engineering is the broader team or function. Quality Assurance is the independent validation activity or role; neither label is a claim of extra staffed headcount.; Open related role or evidence
Continuous Integration (CI)CI: CI means Continuous Integration. Automated source, build, test, security, and evidence checks performed when changes are integrated. CI may block release. CI cannot approve its own exception. A passing CI check does not replace human product ownership or UAT.; Open related role or evidence
Agent Customer-Journey Simulation (M2A)M2A: M2A means Agent Customer-Journey Simulation. A deterministic agent simulation that exercises frozen personas across required product paths. M2A can expose path defects and contradictions, but it cannot establish human understanding.; Open related role or evidence
Agent Customer-Journey Simulation, Round 2 (M2A-R2)M2A-R2: M2A-R2 means Agent Customer-Journey Simulation, Round 2. A deterministic second-round simulation that evaluates frozen personas across required product paths. Official M2A-R2 is not executed and not authorized. It is agent simulation, not human research or human-comprehension evidence.; Open related role or evidence

Explore a product-leadership question

15nodes in system executive view

Full operating model · 8 stops

See how I organize product delivery

System legendTeam functionAgent / workstreamEvidence gateIncident / learningOutcomeCurrentHistoricalFuture / not authorized

Relationship legendAuthorityHandoffBlockerLearningDependencyIndependent reviewAuthorization

Readiness is not a percentage

Seven independent evidence layers

A pass at one layer cannot authorize another. Current blockers stay visible.

01Source readypassed
02Exact deploymentblocked
03Agent Customer-Journey Simulation: Round 2not authorized
04M2H Human Comprehension Researchnot started
05Live AI evidencenot authorized
06External betanot authorized
07Productionnot authorized

Accessible operating model

The complete system, without the canvas.

The full 91-node evidence model remains here regardless of visual density. Every role identity, responsibility, handoff, gate, incident, lesson, contingency and outcome is available without WebGL.

Product mission

MIN–MAX LivingMIN–MAX · MIN–MAX operating system · explicit · currentMIN–MAX Living. operations. Broader team: MIN–MAX operating system. Mission: Help renters make trustworthy apartment decisions. Operating safeguards: Shared release-approval rules apply.
Help renters make trustworthy apartment decisions.
Protects
- Help renters make trustworthy apartment decisions.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Accountable product decision ownership
- Product source reaches the first evidence layer
Remaining blockers
- Environment compatibility
- Human comprehension
- Live-AI evidence
- External beta

Accountable product leadership

Ola Sobo: Accountable Product OrchestratorOla Sobo · Executive Accountability & Risk · explicit · currentOla Sobo: Accountable Product Orchestrator. accountable-owner. Broader team: Executive Accountability & Risk. Mission: Own product strategy, prioritization, risk, validation and release decisions. Operating safeguards: Missing evidence remains unresolved until independently supported; Specialist execution and owner accountability are attributed separately.
Own product strategy, prioritization, risk, validation and release decisions.
Protects
- Own product strategy, prioritization, risk, validation and release decisions.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Set strategy and priorities
- Accept or reject bounded risk
- Authorize merge, deployment, rollback and expansion separately
- Missing evidence remains unresolved until independently supported
- Specialist execution and owner accountability are attributed separately
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Accountable product decision ownership
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights
- Defines the team boundary and decision rights

Product-company functions

Product Leadership / Product ManagementPL · Product Leadership / Product Management · evidence-derived · currentProduct Leadership / Product Management. governance. Broader team: Product Leadership / Product Management. Mission: Frames the renter decision, sequence and value boundary. Operating safeguards: Functional area in the delivery model.
Frames the renter decision, sequence and value boundary.
Protects
- Frames the renter decision, sequence and value boundary.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
Product Engineering / EngineeringPE · Product Engineering / Engineering · evidence-derived · currentProduct Engineering / Engineering. execution. Broader team: Product Engineering / Engineering. Mission: Builds bounded, testable product behavior. Operating safeguards: Functional area in the delivery model.
Builds bounded, testable product behavior.
Protects
- Builds bounded, testable product behavior.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Quality EngineeringQE · Quality Engineering · evidence-derived · currentQuality Engineering. independent-review. Broader team: Quality Engineering. Mission: The broader quality function; Quality Assurance (QA) is its independent validation activity or role. Operating safeguards: Functional area in the delivery model.
The broader quality function; Quality Assurance (QA) is its independent validation activity or role.
Protects
- The broader quality function; Quality Assurance (QA) is its independent validation activity or role.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
User Acceptance & Product OperationsUAT · User Acceptance & Product Operations · evidence-derived · currentUser Acceptance & Product Operations. operations. Broader team: User Acceptance & Product Operations. Mission: Tests complete owner journeys in the real environment; unlike Quality Assurance (QA), it validates end-to-end customer acceptance in the exact product context. Operating safeguards: Functional area in the delivery model.
Tests complete owner journeys in the real environment; unlike Quality Assurance (QA), it validates end-to-end customer acceptance in the exact product context.
Protects
- Tests complete owner journeys in the real environment; unlike Quality Assurance (QA), it validates end-to-end customer acceptance in the exact product context.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
User Research & Human ComprehensionUR · User Research & Human Comprehension · evidence-derived · currentUser Research & Human Comprehension. research. Broader team: User Research & Human Comprehension. Mission: Separates Agent Customer-Journey Simulation (M2A) from M2H evidence created with people. Operating safeguards: Functional area in the delivery model.
Separates Agent Customer-Journey Simulation (M2A) from M2H evidence created with people.
Protects
- Separates Agent Customer-Journey Simulation (M2A) from M2H evidence created with people.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Decision Science and ScoringDS · Decision Science and Scoring · evidence-derived · currentDecision Science and Scoring. execution. Broader team: Decision Science and Scoring. Mission: Protects cost, comparison and scoring integrity. Operating safeguards: Functional area in the delivery model.
Protects cost, comparison and scoring integrity.
Protects
- Protects cost, comparison and scoring integrity.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Platform & Data OperationsPO · Platform & Data Operations · evidence-derived · currentPlatform & Data Operations. operations. Broader team: Platform & Data Operations. Mission: Protects persistence, runtime and environment compatibility. Operating safeguards: Functional area in the delivery model.
Protects persistence, runtime and environment compatibility.
Protects
- Protects persistence, runtime and environment compatibility.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Reusable remediation control
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Security & Privacy GovernanceSG · Security & Privacy Governance · evidence-derived · currentSecurity & Privacy Governance. governance. Broader team: Security & Privacy Governance. Mission: Keeps access, evidence and claims bounded. Operating safeguards: Functional area in the delivery model.
Keeps access, evidence and claims bounded.
Protects
- Keeps access, evidence and claims bounded.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Reusable remediation control
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Release Assurance & Source CustodyRA · Release Assurance & Source Custody · evidence-derived · currentRelease Assurance & Source Custody. governance. Broader team: Release Assurance & Source Custody. Mission: Binds source, tests, deployment and rollback. Operating safeguards: Functional area in the delivery model.
Binds source, tests, deployment and rollback.
Protects
- Binds source, tests, deployment and rollback.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Reusable remediation control
- Role drawn from the recorded workstreams
Accessibility and Inclusive DesignAX · Accessibility and Inclusive Design · evidence-derived · currentAccessibility and Inclusive Design. governance. Broader team: Accessibility and Inclusive Design. Mission: Makes every decision path usable beyond the canvas. Operating safeguards: Functional area in the delivery model.
Makes every decision path usable beyond the canvas.
Protects
- Makes every decision path usable beyond the canvas.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Reusable remediation control
- Role drawn from the recorded workstreams
AI Product and Model EvaluationAI · AI Product and Model Evaluation · evidence-derived · currentAI Product and Model Evaluation. governance. Broader team: AI Product and Model Evaluation. Mission: Keeps live model value measurable and gated. Operating safeguards: Functional area in the delivery model.
Keeps live model value measurable and gated.
Protects
- Keeps live model value measurable and gated.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
Legal, Provider Rights & RiskLR · Legal, Provider Rights & Risk · evidence-derived · currentLegal, Provider Rights & Risk. governance. Broader team: Legal, Provider Rights & Risk. Mission: Protects consent, data rights and permitted use. Operating safeguards: Functional area in the delivery model.
Protects consent, data rights and permitted use.
Protects
- Protects consent, data rights and permitted use.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Beta & Production OperationsBO · Beta & Production Operations · evidence-derived · currentBeta & Production Operations. operations. Broader team: Beta & Production Operations. Mission: Turns readiness evidence into controlled operations. Operating safeguards: Functional area in the delivery model.
Turns readiness evidence into controlled operations.
Protects
- Turns readiness evidence into controlled operations.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams
- Role drawn from the recorded workstreams
Executive Accountability & RiskER · Executive Accountability & Risk · evidence-derived · currentExecutive Accountability & Risk. governance. Broader team: Executive Accountability & Risk. Mission: Retains accountable human stop, rollback and expansion rights. Operating safeguards: Functional area in the delivery model.
Retains accountable human stop, rollback and expansion rights.
Protects
- Retains accountable human stop, rollback and expansion rights.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Functional area in the delivery model
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Defines the team boundary and decision rights
- Role drawn from the recorded workstreams

Canonical roles mapped from audited workstreams

Accountable Product OrchestratorOla · Executive Accountability & Risk · explicit · currentAccountable Product Orchestrator. accountable-owner. Broader team: Executive Accountability & Risk. Mission: Owns strategy, priority, risk acceptance and release decisions. Operating safeguards: Missing evidence remains unresolved until independently supported.
Owns strategy, priority, risk acceptance and release decisions.
Discovered workstream instances (3)
- Apartment Truth Engine owner decisionmerged evidence · Delivered or governed the apartment truth engine owner decision workstream through protected review.
- D1 owner proxyscoped evidence · Carries the bounded d1 owner proxy responsibility in a frozen policy.
- M1 owner proxyscoped evidence · Carries the bounded m1 owner proxy responsibility in a frozen policy.
Protects
- Owns strategy, priority, risk acceptance and release decisions.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Authorize bounded next stages
- Missing evidence remains unresolved until independently supported
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Customer-facing verdict
- Current blockers and reopening conditions
- Separate exact-head owner decision
- Contributes bounded evidence
- Work available for review
- Work available for review
- Work available for review
- Work available for review
Product Strategy and Journey GovernanceJourney · Product Leadership / Product Management · evidence-derived · currentProduct Strategy and Journey Governance. governance. Broader team: Product Leadership / Product Management. Mission: Freezes the coherent renter journey, scenarios and evidence sequence. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Freezes the coherent renter journey, scenarios and evidence sequence.
Discovered workstream instances (3)
- MVP product-gap prioritiesmerged evidence · Delivered or governed the mvp product-gap priorities workstream through protected review.
- MVP owner-subset evidence crosswalkmerged evidence · Delivered or governed the mvp owner-subset evidence crosswalk workstream through protected review.
- MVP user-journey and scenario governancemerged evidence · Delivered or governed the mvp user-journey and scenario governance workstream through protected review.
Protects
- Freezes the coherent renter journey, scenarios and evidence sequence.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Frozen decision and acceptance boundary
- Contributes bounded evidence
- Contributes bounded evidence
- Work available for review
- Work available for review
- Work available for review
Builder / Product Engineering AgentBuilder · Product Engineering / Engineering · explicit · currentBuilder / Product Engineering Agent. execution. Broader team: Product Engineering / Engineering. Mission: Turns bounded product decisions into exact-source candidates. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Turns bounded product decisions into exact-source candidates.
Discovered workstream instances (3)
- WFH trust foundationmerged evidence · Delivered or governed the wfh trust foundation workstream through protected review.
- Version 12 owner-MVP product remediationmerged evidence · Delivered or governed the version 12 owner-mvp product remediation workstream through protected review.
- Final-product UAT defect remediationmerged evidence · Delivered or governed the final-product uat defect remediation workstream through protected review.
Protects
- Turns bounded product decisions into exact-source candidates.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Frozen decision and acceptance boundary
- Reusable remediation control
- Exact-source candidate
- Bounded security review
- Contributes bounded evidence
- Work available for review
- Work available for review
Decision Workspace / Product Experience AgentWorkspace · Product Engineering / Engineering · evidence-derived · currentDecision Workspace / Product Experience Agent. execution. Broader team: Product Engineering / Engineering. Mission: Keeps status, blockers and next actions truthful across surfaces. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Keeps status, blockers and next actions truthful across surfaces.
Discovered workstream instances (2)
- M2A manager-fixture isolationmerged evidence · Delivered or governed the m2a manager-fixture isolation workstream through protected review.
- Decision-first workspace remediationmerged evidence · Delivered or governed the decision-first workspace remediation workstream through protected review.
Protects
- Keeps status, blockers and next actions truthful across surfaces.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Reusable remediation control
- Reusable remediation control
- Reusable remediation control
- Contributes bounded evidence
- Contributes bounded evidence
- Work available for review
- Work available for review
Apartment Truth Engine / Data-Provenance AgentTruth · Platform & Data Operations · explicit · currentApartment Truth Engine / Data-Provenance Agent. execution. Broader team: Platform & Data Operations. Mission: Separates property, unit, quote, source, confidence and authority. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Separates property, unit, quote, source, confidence and authority.
Discovered workstream instances (1)
- Apartment Truth Engine feasibilitymerged evidence · Delivered or governed the apartment truth engine feasibility workstream through protected review.
Protects
- Separates property, unit, quote, source, confidence and authority.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Reusable remediation control
- Reconciled facts and authority
- Contributes bounded evidence
True Cost and Cost-State AgentCost · Decision Science and Scoring · explicit · currentTrue Cost and Cost-State Agent. execution. Broader team: Decision Science and Scoring. Mission: Calculates deterministic cost states without inventing missing values. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Calculates deterministic cost states without inventing missing values.
Discovered workstream instances (2)
- Owner-UAT cost and evidence remediationmerged evidence · Delivered or governed the owner-uat cost and evidence remediation workstream through protected review.
- Apartment-input integrity and scoring gatesmerged evidence · Delivered or governed the apartment-input integrity and scoring gates workstream through protected review.
Protects
- Calculates deterministic cost states without inventing missing values.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Reconciled facts and authority
- Reusable remediation control
- Reusable remediation control
- Cost readiness and blockers
- Contributes bounded evidence
- Work available for review
- Work available for review
Comparison and Like-for-Like AgentCompare · Decision Science and Scoring · evidence-derived · currentComparison and Like-for-Like Agent. execution. Broader team: Decision Science and Scoring. Mission: Blocks false leaders until comparison variance is resolved. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Blocks false leaders until comparison variance is resolved.
Protects
- Blocks false leaders until comparison variance is resolved.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Cost readiness and blockers
- Reusable remediation control
- Valid comparison membership
- Contributes bounded evidence
V3.1 Scoring and Confidence AgentV3.1 · Decision Science and Scoring · evidence-derived · currentV3.1 Scoring and Confidence Agent. execution. Broader team: Decision Science and Scoring. Mission: Keeps Fit, evidence, confidence and readiness distinct. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Keeps Fit, evidence, confidence and readiness distinct.
Discovered workstream instances (3)
- Fit-scale comprehension remediationmerged evidence · Delivered or governed the fit-scale comprehension remediation workstream through protected review.
- Scoring-comprehension remediationmerged evidence · Delivered or governed the scoring-comprehension remediation workstream through protected review.
- Offline recommendation calibrationmerged evidence · Delivered or governed the offline recommendation calibration workstream through protected review.
Protects
- Keeps Fit, evidence, confidence and readiness distinct.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Valid comparison membership
- Shadow model cannot replace authority
V3.2 Shadow Scoring AgentV3.2 · Decision Science and Scoring · explicit · currentV3.2 Shadow Scoring Agent. execution. Broader team: Decision Science and Scoring. Mission: Measures offline drift without changing authoritative scoring. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Measures offline drift without changing authoritative scoring.
Discovered workstream instances (1)
- V3.2 shadow scoring engineactive evidence · Advances shadow scoring evidence without changing the authoritative runtime model.
Protects
- Measures offline drift without changing authoritative scoring.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Reusable remediation control
- Shadow model cannot replace authority
- Contributes bounded evidence
- Work available for review
Product Integrity AgentIntegrity · Quality Engineering · explicit · currentProduct Integrity Agent. independent-review. Broader team: Quality Engineering. Mission: Uses Continuous Integration (CI) to automate source, build, test, security and evidence checks when changes are integrated. Operating safeguards: CI exceptions require independent approval; Human product ownership and UAT remain separate from CI results. Related comprehension concepts: Continuous Integration (CI). Definition: Automated source, build, test, security, and evidence checks performed when changes are integrated. Decision/evidence boundary: CI may block release. CI cannot approve its own exception. A passing CI check does not replace human product ownership or UAT. Concept: not a role, agent, employee or headcount record.
Uses Continuous Integration (CI) to automate source, build, test, security and evidence checks when changes are integrated.
Discovered workstream instances (1)
- Functional QA and security hardeningmerged evidence · Delivered or governed the functional qa and security hardening workstream through protected review.
Protects
- Continuous Integration (CI). Definition: Automated source, build, test, security, and evidence checks performed when changes are integrated. Decision/evidence boundary: CI may block release. CI cannot approve its own exception. A passing CI check does not replace human product ownership or UAT. Concept: not a role, agent, employee or headcount record.
- Exact source, build, test, security and evidence integrity
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: A Continuous Integration (CI) result bound to the exact candidate
Scope and safeguards
- Block integration when an automated source, build, test, security or evidence gate fails
- CI exceptions require independent approval
- Human product ownership and UAT remain separate from CI results
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Exact-source candidate
- Exact-head evidence
- Contributes bounded evidence
- Work available for review
Security Baseline AgentSecurity · Security & Privacy Governance · explicit · currentSecurity Baseline Agent. independent-review. Broader team: Security & Privacy Governance. Mission: Runs source, dependency, access and secret controls. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Runs source, dependency, access and secret controls.
Discovered workstream instances (1)
- External-access safety gatemerged evidence · Delivered or governed the external-access safety gate workstream through protected review.
Protects
- Runs source, dependency, access and secret controls.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Bounded security review
- Independent security result
- Contributes bounded evidence
Independent Exact-Head Quality Assurance (QA)QA · Quality Engineering · explicit · currentIndependent Exact-Head Quality Assurance (QA). independent-review. Broader team: Quality Engineering. Mission: Quality Assurance (QA) independently validates exact source and intended behavior without inheriting Builder conclusions. Operating safeguards: Candidate changes are reported separately to preserve QA independence; Test defects remain distinct from product defects. Related comprehension concepts: Test defect. Definition: The test expectation, fixture, locator, or assertion is incorrect even though the intended product behavior works. Example: A diagnostic expected the message field’s accessible name to equal one exact string even though the field was correctly labelled and described. Customer consequence: Delivery is blocked by faulty evidence rather than faulty product behavior. Remediation boundary: May require test remediation. Concept: not a role, agent, employee or headcount record. Quality Assurance (QA). Definition: The independent validation activity or role that challenges source and intended product behavior. Decision/evidence boundary: Quality Engineering is the broader team or function. Quality Assurance is the independent validation activity or role; neither label is a claim of extra staffed headcount. Concept: not a role, agent, employee or headcount record.
Quality Assurance (QA) independently validates exact source and intended behavior without inheriting Builder conclusions.
Protects
- Test defect. Definition: The test expectation, fixture, locator, or assertion is incorrect even though the intended product behavior works. Example: A diagnostic expected the message field’s accessible name to equal one exact string even though the field was correctly labelled and described. Customer consequence: Delivery is blocked by faulty evidence rather than faulty product behavior. Remediation boundary: May require test remediation. Concept: not a role, agent, employee or headcount record.
- Quality Assurance (QA). Definition: The independent validation activity or role that challenges source and intended product behavior. Decision/evidence boundary: Quality Engineering is the broader team or function. Quality Assurance is the independent validation activity or role; neither label is a claim of extra staffed headcount. Concept: not a role, agent, employee or headcount record.
- Independent exact-head Quality Assurance (QA) judgment
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: An independent source-and-behavior verdict with exact evidence custody
Scope and safeguards
- Block advancement when exact source or intended product behavior fails
- Candidate changes are reported separately to preserve QA independence
- Test defects remain distinct from product defects
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Exact-head evidence
- Independent security result
- Merge-safe candidate for deployed review
- Finding and customer consequence
- Finding and customer consequence
- Contributes bounded evidence
- Contributes bounded evidence
- Work available for review
Customer-Trust Adversarial QA AgentTrust QA · Quality Engineering · evidence-derived · currentCustomer-Trust Adversarial QA Agent. independent-review. Broader team: Quality Engineering. Mission: Finds contradictions, dead ends and cross-surface drift. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Finds contradictions, dead ends and cross-surface drift.
Discovered workstream instances (2)
- Version 14 owner-trust remediationmerged evidence · Delivered or governed the version 14 owner-trust remediation workstream through protected review.
- Version 9 deployed-integrity remediationmerged evidence · Delivered or governed the version 9 deployed-integrity remediation workstream through protected review.
Protects
- Finds contradictions, dead ends and cross-surface drift.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Finding and customer consequence
- Finding and customer consequence
- Contributes bounded evidence
- Work available for review
User Acceptance & Product Operations AgentUAT · User Acceptance & Product Operations · explicit · currentUser Acceptance & Product Operations Agent. operations. Broader team: User Acceptance & Product Operations. Mission: Exercises complete authenticated owner journeys after independent QA and preserves acceptance and cleanup evidence. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Exercises complete authenticated owner journeys after independent QA and preserves acceptance and cleanup evidence.
Discovered workstream instances (3)
- Version 12 blocked-path evidence controlsmerged evidence · Delivered or governed the version 12 blocked-path evidence controls workstream through protected review.
- Owner-UAT governancemerged evidence · Delivered or governed the owner-uat governance workstream through protected review.
- Disabled-foundation owner-UAT evidencemerged evidence · Delivered or governed the disabled-foundation owner-uat evidence workstream through protected review.
Protects
- Exercises complete authenticated owner journeys after independent QA and preserves acceptance and cleanup evidence.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Merge-safe candidate for deployed review
- Customer-facing verdict
- Finding and customer consequence
- Finding and customer consequence
- Finding and customer consequence
- Finding and customer consequence
Agent Customer-Journey Simulation: Round 2M2A-R2 · User Research & Human Comprehension · explicit · futureAgent Customer-Journey Simulation: Round 2. research. Broader team: User Research & Human Comprehension. Mission: A deterministic second-round Agent Customer-Journey Simulation (M2A) that evaluates frozen personas across required product paths without claiming human evidence. Operating safeguards: Human validation and completed user research require human evidence; Official M2A-R2 execution requires separate authorization. Related comprehension concepts: Agent Customer-Journey Simulation (M2A). Definition: A deterministic agent simulation that exercises frozen personas across required product paths. Decision/evidence boundary: M2A can expose path defects and contradictions, but it cannot establish human understanding. Concept: not a role, agent, employee or headcount record. Agent Customer-Journey Simulation, Round 2 (M2A-R2). Definition: A deterministic second-round simulation that evaluates frozen personas across required product paths. Decision/evidence boundary: Official M2A-R2 is not executed and not authorized. It is agent simulation, not human research or human-comprehension evidence. Concept: not a role, agent, employee or headcount record.
A deterministic second-round Agent Customer-Journey Simulation (M2A) that evaluates frozen personas across required product paths without claiming human evidence.
Discovered workstream instances (2)
- Version 11 M2A execution packagemerged evidence · Delivered or governed the version 11 m2a execution package workstream through protected review.
- M2A exact-path evidence harnessmerged evidence · Delivered or governed the m2a exact-path evidence harness workstream through protected review.
Protects
- Agent Customer-Journey Simulation (M2A). Definition: A deterministic agent simulation that exercises frozen personas across required product paths. Decision/evidence boundary: M2A can expose path defects and contradictions, but it cannot establish human understanding. Concept: not a role, agent, employee or headcount record.
- Agent Customer-Journey Simulation, Round 2 (M2A-R2). Definition: A deterministic second-round simulation that evaluates frozen personas across required product paths. Decision/evidence boundary: Official M2A-R2 is not executed and not authorized. It is agent simulation, not human research or human-comprehension evidence. Concept: not a role, agent, employee or headcount record.
- Deterministic agent-simulation evidence kept separate from human evidence
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: A frozen persona-by-path Agent Customer-Journey Simulation: Round 2 result
Scope and safeguards
- Expose deterministic path defects and contradictions when separately authorized
- Human validation and completed user research require human evidence
- Official M2A-R2 execution requires separate authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Accepted environment required first
- Simulation does not replace human evidence
- Contributes bounded evidence
- Work available for review
- Work available for review
Remaining blockers
- Official M2A-R2 is not executed and not authorized
M2H Human Comprehension Research / Research OperationsM2H · User Research & Human Comprehension · evidence-derived · futureM2H Human Comprehension Research / Research Operations. research. Broader team: User Research & Human Comprehension. Mission: Owns future human-comprehension study gates and stop decisions that cannot be inferred from simulation. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Owns future human-comprehension study gates and stop decisions that cannot be inferred from simulation.
Discovered workstream instances (3)
- M1 and M2H owner-proxy preparationmerged evidence · Delivered or governed the m1 and m2h owner-proxy preparation workstream through protected review.
- M2H precontact operating packagemerged evidence · Delivered or governed the m2h precontact operating package workstream through protected review.
- M2H study sponsor and owner proxyscoped evidence · Carries the bounded m2h study sponsor and owner proxy responsibility in a frozen policy.
Protects
- Owns future human-comprehension study gates and stop decisions that cannot be inferred from simulation.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Simulation does not replace human evidence
- Human and live-AI lanes remain separate
- Contributes bounded evidence
Recruitment CoordinatorRecruit · User Research & Human Comprehension · explicit · futureRecruitment Coordinator. research. Broader team: User Research & Human Comprehension. Mission: Keeps future contact records separate from research evidence. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Keeps future contact records separate from research evidence.
Discovered workstream instances (1)
- M2H recruitment coordinatorscoped evidence · Carries the bounded m2h recruitment coordinator responsibility in a frozen policy.
Protects
- Keeps future contact records separate from research evidence.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
Human ModeratorModerator · User Research & Human Comprehension · explicit · futureHuman Moderator. research. Broader team: User Research & Human Comprehension. Mission: Uses a neutral script and records first responses before probing. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Uses a neutral script and records first responses before probing.
Discovered workstream instances (1)
- M2H moderatorscoped evidence · Carries the bounded m2h moderator responsibility in a frozen policy.
Protects
- Uses a neutral script and records first responses before probing.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Contributes bounded evidence
- Work available for review
Independent Research AnalystAnalyst · User Research & Human Comprehension · explicit · futureIndependent Research Analyst. research. Broader team: User Research & Human Comprehension. Mission: Codes pseudonymous observations without contact-data access. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Codes pseudonymous observations without contact-data access.
Discovered workstream instances (1)
- M2H analystscoped evidence · Carries the bounded m2h analyst responsibility in a frozen policy.
Protects
- Codes pseudonymous observations without contact-data access.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Contributes bounded evidence
- Work available for review
Independent AdjudicatorAdjudicator · Security & Privacy Governance · explicit · currentIndependent Adjudicator. independent-review. Broader team: Security & Privacy Governance. Mission: Recomputes gates without operating the run being judged. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Recomputes gates without operating the run being judged.
Discovered workstream instances (2)
- M1 independent adjudicatorscoped evidence · Carries the bounded m1 independent adjudicator responsibility in a frozen policy.
- M2H independent adjudicatorscoped evidence · Carries the bounded m2h independent adjudicator responsibility in a frozen policy.
Protects
- Recomputes gates without operating the run being judged.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Finding and customer consequence
- Contributes bounded evidence
- Work available for review
- Work available for review
X-05 Accessibility and Viewport AgentX-05 · Accessibility and Inclusive Design · explicit · currentX-05 Accessibility and Viewport Agent. governance. Broader team: Accessibility and Inclusive Design. Mission: Keeps emulation, exact-source automation and physical corroboration distinct. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Keeps emulation, exact-source automation and physical corroboration distinct.
Discovered workstream instances (2)
- Version 14 X-05 trust governancemerged evidence · Delivered or governed the version 14 x-05 trust governance workstream through protected review.
- M2H accessibility coordinatorscoped evidence · Carries the bounded m2h accessibility coordinator responsibility in a frozen policy.
Protects
- Keeps emulation, exact-source automation and physical corroboration distinct.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Three-layer viewport evidence not executed
- Finding and customer consequence
- Work available for review
Capability and Risk Governance AgentGovernance · Security & Privacy Governance · explicit · currentCapability and Risk Governance Agent. governance. Broader team: Security & Privacy Governance. Mission: Maintains independent readiness dimensions and reopening conditions. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Maintains independent readiness dimensions and reopening conditions.
Discovered workstream instances (6)
- Version 12 non-live preparationmerged evidence · Delivered or governed the version 12 non-live preparation workstream through protected review.
- Post-harness capability-risk governancemerged evidence · Delivered or governed the post-harness capability-risk governance workstream through protected review.
- Capability-readiness governancemerged evidence · Delivered or governed the capability-readiness governance workstream through protected review.
- Planning-governance packagemerged evidence · Delivered or governed the planning-governance package workstream through protected review.
- Authorized-enrichment governance clarificationmerged evidence · Delivered or governed the authorized-enrichment governance clarification workstream through protected review.
- Live-provider governance recordmerged evidence · Delivered or governed the live-provider governance record workstream through protected review.
Protects
- Maintains independent readiness dimensions and reopening conditions.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Current blockers and reopening conditions
- Contributes bounded evidence
- Work available for review
- Work available for review
Source-Custody / Release AgentCustody · Release Assurance & Source Custody · explicit · currentSource-Custody / Release Agent. independent-review. Broader team: Release Assurance & Source Custody. Mission: Binds commit, tree, application, workflows, deployment and rollback. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Binds commit, tree, application, workflows, deployment and rollback.
Discovered workstream instances (3)
- Authenticated QA build provenancemerged evidence · Delivered or governed the authenticated qa build provenance workstream through protected review.
- Main-governance recordmerged evidence · Delivered or governed the main-governance record workstream through protected review.
- Branch-protection verificationmerged evidence · Delivered or governed the branch-protection verification workstream through protected review.
Protects
- Binds commit, tree, application, workflows, deployment and rollback.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Separate exact-head owner decision
- Protected merge and deployment custody
- Finding and customer consequence
- Contributes bounded evidence
- Work available for review
- Work available for review
- Work available for review
D1 Read and Environment-Reconciliation AgentD1 · Platform & Data Operations · explicit · currentD1 Read and Environment-Reconciliation Agent. operations. Broader team: Platform & Data Operations. Mission: Reconciles environment identity and schema using bounded read-only evidence. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Reconciles environment identity and schema using bounded read-only evidence.
Discovered workstream instances (5)
- Owner-only D1 inspection routemerged evidence · Delivered or governed the owner-only d1 inspection route workstream through protected review.
- D1 owner-proxy recoverymerged evidence · Delivered or governed the d1 owner-proxy recovery workstream through protected review.
- D1 Read runner preparationmerged evidence · Delivered or governed the d1 read runner preparation workstream through protected review.
- Version 13 D1 and X-05 reconciliationmerged evidence · Delivered or governed the version 13 d1 and x-05 reconciliation workstream through protected review.
- D1 Read execution operatorscoped evidence · Carries the bounded d1 read execution operator responsibility in a frozen policy.
Protects
- Reconciles environment identity and schema using bounded read-only evidence.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Future least-privilege secure input
- Independent reconciliation review
- Schema compatibility unresolved
- Finding and customer consequence
- Work available for review
- Work available for review
Credential Custodian / Revocation AgentCredential · Security & Privacy Governance · explicit · currentCredential Custodian / Revocation Agent. governance. Broader team: Security & Privacy Governance. Mission: Keeps ephemeral credentials out of evidence and proves revocation. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Keeps ephemeral credentials out of evidence and proves revocation.
Discovered workstream instances (3)
- D1 credential and revocation custodianscoped evidence · Carries the bounded d1 credential and revocation custodian responsibility in a frozen policy.
- M1 credential custodianscoped evidence · Carries the bounded m1 credential custodian responsibility in a frozen policy.
- M1 kill-switch operatorscoped evidence · Carries the bounded m1 kill-switch operator responsibility in a frozen policy.
Protects
- Keeps ephemeral credentials out of evidence and proves revocation.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Future least-privilege secure input
- Contributes bounded evidence
Evidence Reviewer / Custody AgentEvidence · Security & Privacy Governance · evidence-derived · currentEvidence Reviewer / Custody Agent. independent-review. Broader team: Security & Privacy Governance. Mission: Checks provenance, privacy, retention and cleanup independently. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Checks provenance, privacy, retention and cleanup independently.
Discovered workstream instances (3)
- Authenticated evidence-recovery decisionmerged evidence · Delivered or governed the authenticated evidence-recovery decision workstream through protected review.
- Authenticated-UAT bounded owner decisionmerged evidence · Delivered or governed the authenticated-uat bounded owner decision workstream through protected review.
- D1 independent evidence reviewerscoped evidence · Carries the bounded d1 independent evidence reviewer responsibility in a frozen policy.
Protects
- Checks provenance, privacy, retention and cleanup independently.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Independent reconciliation review
- Finding and customer consequence
- Contributes bounded evidence
- Work available for review
M1 Live-AI Readiness / Model-Evaluation AgentM1 · AI Product and Model Evaluation · explicit · futureM1 Live-AI Readiness / Model-Evaluation Agent. governance. Broader team: AI Product and Model Evaluation. Mission: Keeps a frozen live-model benchmark disabled until separately authorized. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Keeps a frozen live-model benchmark disabled until separately authorized.
Discovered workstream instances (6)
- M1 owner-decision packagemerged evidence · Delivered or governed the m1 owner-decision package workstream through protected review.
- Deterministic live-candidate test clockmerged evidence · Delivered or governed the deterministic live-candidate test clock workstream through protected review.
- Live-activation candidate controlsmerged evidence · Delivered or governed the live-activation candidate controls workstream through protected review.
- Evidence-interpreter preparationmerged evidence · Delivered or governed the evidence-interpreter preparation workstream through protected review.
- M1 model-candidate reviewerscoped evidence · Carries the bounded m1 model-candidate reviewer responsibility in a frozen policy.
- M1 future benchmark operatorscoped evidence · Carries the bounded m1 future benchmark operator responsibility in a frozen policy.
Protects
- Keeps a frozen live-model benchmark disabled until separately authorized.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Contributes bounded evidence
- Work available for review
Provider-Rights and Data-Enrichment AgentProvider · Legal, Provider Rights & Risk · explicit · currentProvider-Rights and Data-Enrichment Agent. governance. Broader team: Legal, Provider Rights & Risk. Mission: Evaluates source rights, entity scope, accuracy, cost and kill switches. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Evaluates source rights, entity scope, accuracy, cost and kill switches.
Discovered workstream instances (4)
- Provider owner-decision packagemerged evidence · Delivered or governed the provider owner-decision package workstream through protected review.
- Provider stage owner-decision packagemerged evidence · Delivered or governed the provider stage owner-decision package workstream through protected review.
- Authorized-data-enrichment pilotmerged evidence · Delivered or governed the authorized-data-enrichment pilot workstream through protected review.
- Live-provider pilot foundationmerged evidence · Delivered or governed the live-provider pilot foundation workstream through protected review.
Protects
- Evaluates source rights, entity scope, accuracy, cost and kill switches.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Live provider rights incomplete
- Work available for review
Legal, Privacy and Fair-Housing Review AgentLegal · Legal, Provider Rights & Risk · evidence-derived · currentLegal, Privacy and Fair-Housing Review Agent. governance. Broader team: Legal, Provider Rights & Risk. Mission: Reviews consent, sharing, protected-data proxies and source rights. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Reviews consent, sharing, protected-data proxies and source rights.
Discovered workstream instances (2)
- Provider and legal governance recordmerged evidence · Delivered or governed the provider and legal governance record workstream through protected review.
- M2H privacy and cleanup operatorscoped evidence · Carries the bounded m2h privacy and cleanup operator responsibility in a frozen policy.
Protects
- Reviews consent, sharing, protected-data proxies and source rights.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Signed legal and privacy evidence incomplete
- Contributes bounded evidence
External-Beta Readiness AgentBeta · Beta & Production Operations · explicit · futureExternal-Beta Readiness Agent. operations. Broader team: Beta & Production Operations. Mission: Integrates privacy, support, recovery, rights and metric gates. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Integrates privacy, support, recovery, rights and metric gates.
Discovered workstream instances (1)
- External-beta metrics planmerged evidence · Delivered or governed the external-beta metrics plan workstream through protected review.
Protects
- Integrates privacy, support, recovery, rights and metric gates.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Contributes bounded evidence
- Contributes bounded evidence
- Work available for review
Production Operations / Incident AgentProd Ops · Beta & Production Operations · evidence-derived · futureProduction Operations / Incident Agent. operations. Broader team: Beta & Production Operations. Mission: Owns rollback, severity response, support and post-incident learning. Operating safeguards: Merge, deployment, beta and production require explicit owner authorization.
Owns rollback, severity response, support and post-incident learning.
Discovered workstream instances (6)
- Domain and launch-operations planmerged evidence · Delivered or governed the domain and launch-operations plan workstream through protected review.
- Separate QA and restricted production environmentsmerged evidence · Delivered or governed the separate qa and restricted production environments workstream through protected review.
- Storage-escalation recordmerged evidence · Delivered or governed the storage-escalation record workstream through protected review.
- Staging recoverymerged evidence · Delivered or governed the staging recovery workstream through protected review.
- M1 incident commanderscoped evidence · Carries the bounded m1 incident commander responsibility in a frozen policy.
- M2H incident ownerscoped evidence · Carries the bounded m2h incident owner responsibility in a frozen policy.
Protects
- Owns rollback, severity response, support and post-incident learning.
What it uses and produces
- Uses: Bounded product scope
- Uses: Exact-source or evidence contract
- Produces: Reviewable source, evidence, finding or decision input
Scope and safeguards
- Act only inside the named functional boundary
- Merge, deployment, beta and production require explicit owner authorization
Primary handoffs
- Hands exact, reviewable output to the next independent role
Handoffs and relationships
- Role drawn from the recorded workstreams
- Contributes bounded evidence
- Contributes bounded evidence
- Work available for review

Independent evidence gates

Source readyMIN–MAX operating system · explicit · currentSource ready. independent-review. Broader team: MIN–MAX operating system. Mission: Protected source integrates current trust remediations. Operating safeguards: Each evidence layer requires its own pass.
Protected source integrates current trust remediations.
Protects
- Protected source integrates current trust remediations.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Product source reaches the first evidence layer
- A source pass does not prove deployed behavior
Exact deploymentMIN–MAX operating system · explicit · currentExact deployment. independent-review. Broader team: MIN–MAX operating system. Mission: Version 13 is deployed but not accepted; environment compatibility remains unresolved. Operating safeguards: Each evidence layer requires its own pass.
Version 13 is deployed but not accepted; environment compatibility remains unresolved.
Protects
- Version 13 is deployed but not accepted; environment compatibility remains unresolved.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Protected merge and deployment custody
- A source pass does not prove deployed behavior
- Schema compatibility unresolved
- Three-layer viewport evidence not executed
- Accepted deployment precedes official simulation
- Accepted environment required first
Remaining blockers
- Version 13 is deployed but not accepted; environment compatibility remains unresolved.
Agent Customer-Journey Simulation: Round 2MIN–MAX operating system · explicit · futureAgent Customer-Journey Simulation: Round 2. independent-review. Broader team: MIN–MAX operating system. Mission: Official M2A-R2 is not executed and not authorized; this agent simulation is not human evidence. Operating safeguards: Each evidence layer requires its own pass.
Official M2A-R2 is not executed and not authorized; this agent simulation is not human evidence.
Protects
- Official M2A-R2 is not executed and not authorized; this agent simulation is not human evidence.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Accepted deployment precedes official simulation
- Simulation cannot substitute for human evidence
Remaining blockers
- Official M2A-R2 is not executed and not authorized; this agent simulation is not human evidence.
M2H Human Comprehension ResearchMIN–MAX operating system · explicit · futureM2H Human Comprehension Research. independent-review. Broader team: MIN–MAX operating system. Mission: No participant activity or human verdict exists; this evidence cannot be inferred from M2A. Operating safeguards: Each evidence layer requires its own pass.
No participant activity or human verdict exists; this evidence cannot be inferred from M2A.
Protects
- No participant activity or human verdict exists; this evidence cannot be inferred from M2A.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Simulation cannot substitute for human evidence
- Human and model evidence remain separate
Remaining blockers
- No participant activity or human verdict exists; this evidence cannot be inferred from M2A.
Live AI evidenceMIN–MAX operating system · explicit · futureLive AI evidence. independent-review. Broader team: MIN–MAX operating system. Mission: Model, project, billing, credential and live calls remain absent. Operating safeguards: Each evidence layer requires its own pass.
Model, project, billing, credential and live calls remain absent.
Protects
- Model, project, billing, credential and live calls remain absent.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Human and model evidence remain separate
- Human and live-AI lanes remain separate
- Bounded live value required
Remaining blockers
- Model, project, billing, credential and live calls remain absent.
External betaMIN–MAX operating system · explicit · futureExternal beta. independent-review. Broader team: MIN–MAX operating system. Mission: Current decision is NO-GO. Operating safeguards: Each evidence layer requires its own pass.
Current decision is NO-GO.
Protects
- Current decision is NO-GO.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Bounded live value required
- Live provider rights incomplete
- Signed legal and privacy evidence incomplete
- Evidence-authorized expansion
Remaining blockers
- Current decision is NO-GO.
ProductionMIN–MAX operating system · explicit · futureProduction. independent-review. Broader team: MIN–MAX operating system. Mission: Production remains unchanged and out of scope. Operating safeguards: Each evidence layer requires its own pass.
Production remains unchanged and out of scope.
Protects
- Production remains unchanged and out of scope.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Each evidence layer requires its own pass.
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Evidence-authorized expansion
Remaining blockers
- Production remains unchanged and out of scope.

Incidents, lessons and contingencies

Unknown presented as zeroMIN–MAX operating system · explicit · historicalUnknown presented as zero. governance. Broader team: MIN–MAX operating system. Mission: A missing fact appeared certain. Operating safeguards: Shared release-approval rules apply.
A missing fact appeared certain.
Protects
- A missing fact appeared certain.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Unknown/zero regression fixtures. Historical source defect closed; future deployment proof remains.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
Stale quote without actionMIN–MAX operating system · explicit · historicalStale quote without action. governance. Broader team: MIN–MAX operating system. Mission: A stale quote had no recovery action. Operating safeguards: Shared release-approval rules apply.
A stale quote had no recovery action.
Protects
- A stale quote had no recovery action.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Executable freshness control. Closed in targeted evidence; rerun on the next exact deployment.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
Like-for-like bypassMIN–MAX operating system · explicit · historicalLike-for-like bypass. governance. Broader team: MIN–MAX operating system. Mission: Unlike homes could produce a leader. Operating safeguards: Shared release-approval rules apply.
Unlike homes could produce a leader.
Protects
- Unlike homes could produce a leader.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Membership-bound variance control. Historical blocker closed in source; deployment rerun required.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
Zero-active comparison contradictionMIN–MAX operating system · explicit · historicalZero-active comparison contradiction. governance. Broader team: MIN–MAX operating system. Mission: The interface claimed an active comparison with no active set. Operating safeguards: Shared release-approval rules apply.
The interface claimed an active comparison with no active set.
Protects
- The interface claimed an active comparison with no active set.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Canonical state plus signed transient feedback. Historical failed deployment preserved; later source corrected.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
- Work available for review
Inclusive total treated as fees-onlyMIN–MAX operating system · explicit · historicalInclusive total treated as fees-only. governance. Broader team: MIN–MAX operating system. Mission: A confirmed inclusive price was presented with the wrong basis. Operating safeguards: Shared release-approval rules apply.
A confirmed inclusive price was presented with the wrong basis.
Protects
- A confirmed inclusive price was presented with the wrong basis.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Fail-closed cost/presentation boundary. Historical failure closed in source; deployment pending.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
WFH dead-end actionMIN–MAX operating system · explicit · historicalWFH dead-end action. Product defect. The product behaves incorrectly, contradicts its intended decision model, creates a dead end, or presents an unsafe or misleading customer state. Customer consequence: The visitor reaches a dead end or receives an untrustworthy decision. Remediation boundary: May require product remediation. Related comprehension concepts: Product defect. Definition: The product behaves incorrectly, contradicts its intended decision model, creates a dead end, or presents an unsafe or misleading customer state. Example: A page displays “Review work-from-home requirement” even though the environment cannot execute the work-from-home question. Customer consequence: The visitor reaches a dead end or receives an untrustworthy decision. Remediation boundary: May require product remediation. Concept: not a role, agent, employee or headcount record.
The product advertised an unavailable assessment.
Protects
- Product defect. Definition: The product behaves incorrectly, contradicts its intended decision model, creates a dead end, or presents an unsafe or misleading customer state. Example: A page displays “Review work-from-home requirement” even though the environment cannot execute the work-from-home question. Customer consequence: The visitor reaches a dead end or receives an untrustworthy decision. Remediation boundary: May require product remediation. Concept: not a role, agent, employee or headcount record.
- The product advertised an unavailable assessment.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Action-availability contract. Closed in protected source; exact deployment pending.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
- Work available for review
Shared check-count divergenceMIN–MAX operating system · explicit · historicalShared check-count divergence. governance. Broader team: MIN–MAX operating system. Mission: Owner and shared reports could disagree on readiness. Operating safeguards: Shared release-approval rules apply.
Owner and shared reports could disagree on readiness.
Protects
- Owner and shared reports could disagree on readiness.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Legacy counts fail closed. Closed in protected source; exact deployment pending.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
- Work available for review
Version 12 product failuresMIN–MAX operating system · explicit · historicalVersion 12 product failures. governance. Broader team: MIN–MAX operating system. Mission: Setup, variance, risk copy and mobile status diverged. Operating safeguards: Shared release-approval rules apply.
Setup, variance, risk copy and mobile status diverged.
Protects
- Setup, variance, risk copy and mobile status diverged.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Immutable failed evidence plus regression suite. Version 12 remains failed; its source lessons were integrated.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
- Work available for review
D1 schema incompatibilityMIN–MAX operating system · explicit · currentD1 schema incompatibility. Environment defect. The deployed or test environment is incompatible with the approved source, schema, runtime, access, viewport, or binding contract. Customer consequence: Product behavior remains unproven in that environment. Remediation boundary: Requires environment reconciliation or a new environment. Related comprehension concepts: Environment defect. Definition: The deployed or test environment is incompatible with the approved source, schema, runtime, access, viewport, or binding contract. Example: A QA database lacked required foundational tables, so the customer journey could not be executed end to end. Customer consequence: Product behavior remains unproven in that environment. Remediation boundary: Requires environment reconciliation or a new environment. Concept: not a role, agent, employee or headcount record.
Only 50 of 56 expected tables were visible.
Protects
- Environment defect. Definition: The deployed or test environment is incompatible with the approved source, schema, runtime, access, viewport, or binding contract. Example: A QA database lacked required foundational tables, so the customer journey could not be executed end to end. Customer consequence: Product behavior remains unproven in that environment. Remediation boundary: Requires environment reconciliation or a new environment. Concept: not a role, agent, employee or headcount record.
- Only 50 of 56 expected tables were visible.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Credential-free D1 read runner. Current blocker: root cause remains unresolved.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
- Work available for review
Remaining blockers
- Current blocker: root cause remains unresolved.
Authenticated evidence limitationMIN–MAX operating system · explicit · currentAuthenticated evidence limitation. Evidence-harness defect. The mechanism used to create or retain evidence fails before it can validly judge the product. Customer consequence: The product remains unadjudicated; no pass or failure may be inferred. Remediation boundary: Requires valid evidence before judging the product. Related comprehension concepts: Evidence-harness defect. Definition: The mechanism used to create or retain evidence fails before it can validly judge the product. Example: An evidence package could not start because its output directory or attestation-output contract was invalid. Customer consequence: The product remains unadjudicated; no pass or failure may be inferred. Remediation boundary: Requires valid evidence before judging the product. Concept: not a role, agent, employee or headcount record.
A normal authenticated capture could not be established.
Protects
- Evidence-harness defect. Definition: The mechanism used to create or retain evidence fails before it can validly judge the product. Example: An evidence package could not start because its output directory or attestation-output contract was invalid. Customer consequence: The product remains unadjudicated; no pass or failure may be inferred. Remediation boundary: Requires valid evidence before judging the product. Concept: not a role, agent, employee or headcount record.
- A normal authenticated capture could not be established.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Authenticated-evidence contract. Current evidence limitation; not a product pass.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
Remaining blockers
- Current evidence limitation; not a product pass.
Viewport evidence limitationMIN–MAX operating system · explicit · currentViewport evidence limitation. governance. Broader team: MIN–MAX operating system. Mission: A mobile claim lacked correctly labelled proof. Operating safeguards: Shared release-approval rules apply.
A mobile claim lacked correctly labelled proof.
Protects
- A mobile claim lacked correctly labelled proof.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Three-layer evidence contract. Method approved; execution has not run.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
- Work available for review
Remaining blockers
- Method approved; execution has not run.
Source-custody driftMIN–MAX operating system · explicit · historicalSource-custody drift. Continuous-integration defect. The automated integration pipeline, test runner, or build orchestration changes or destabilizes the evidence environment. Customer consequence: Release evidence cannot be trusted until the pipeline behavior is corrected. Remediation boundary: Requires pipeline or test-runtime remediation. Related comprehension concepts: Continuous-integration defect. Definition: The automated integration pipeline, test runner, or build orchestration changes or destabilizes the evidence environment. Example: Automatic Continuous Integration (CI) Git-diff capture changed the checkout to a shallow repository after browser tests, invalidating later source-custody checks. Customer consequence: Release evidence cannot be trusted until the pipeline behavior is corrected. Remediation boundary: Requires pipeline or test-runtime remediation. Concept: not a role, agent, employee or headcount record.
A deployed source relationship was not reproducible.
Protects
- Continuous-integration defect. Definition: The automated integration pipeline, test runner, or build orchestration changes or destabilizes the evidence environment. Example: Automatic Continuous Integration (CI) Git-diff capture changed the checkout to a shallow repository after browser tests, invalidating later source-custody checks. Customer consequence: Release evidence cannot be trusted until the pipeline behavior is corrected. Remediation boundary: Requires pipeline or test-runtime remediation. Concept: not a role, agent, employee or headcount record.
- A deployed source relationship was not reproducible.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Exact-source provenance gate. Historical incident closed by later custody controls.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
V3.2 threshold missMIN–MAX operating system · explicit · currentV3.2 threshold miss. governance. Broader team: MIN–MAX operating system. Mission: Shadow rankings exceeded selected drift limits. Operating safeguards: Shared release-approval rules apply.
Shadow rankings exceeded selected drift limits.
Protects
- Shadow rankings exceeded selected drift limits.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Learning
Non-disruptive shadow gate. Current activation blocker, not a production defect.
Handoffs and relationships
- Finding and customer consequence
- Reusable remediation control
Remaining blockers
- Current activation blocker, not a production defect.

My product leadership in practice

Strategic judgmentMIN–MAX operating system · evidence-derived · currentStrategic judgment. governance. Broader team: MIN–MAX operating system. Mission: Sequences customer value, trust and technical dependencies. Operating safeguards: Shared release-approval rules apply.
Sequences customer value, trust and technical dependencies.
Protects
- Sequences customer value, trust and technical dependencies.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
First-principles framingMIN–MAX operating system · evidence-derived · currentFirst-principles framing. governance. Broader team: MIN–MAX operating system. Mission: Starts with the renter decision and consequence, not a requested feature. Operating safeguards: Shared release-approval rules apply.
Starts with the renter decision and consequence, not a requested feature.
Protects
- Starts with the renter decision and consequence, not a requested feature.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Systems thinkingMIN–MAX operating system · evidence-derived · currentSystems thinking. governance. Broader team: MIN–MAX operating system. Mission: Separates states, evidence layers, exceptions and downstream effects. Operating safeguards: Shared release-approval rules apply.
Separates states, evidence layers, exceptions and downstream effects.
Protects
- Separates states, evidence layers, exceptions and downstream effects.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Technical product fluencyMIN–MAX operating system · evidence-derived · currentTechnical product fluency. governance. Broader team: MIN–MAX operating system. Mission: Works across contracts, scoring, persistence, testing and provenance. Operating safeguards: Shared release-approval rules apply.
Works across contracts, scoring, persistence, testing and provenance.
Protects
- Works across contracts, scoring, persistence, testing and provenance.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Cross-functional orchestrationMIN–MAX operating system · evidence-derived · currentCross-functional orchestration. governance. Broader team: MIN–MAX operating system. Mission: Connects product, engineering, QA, research, security and platform. Operating safeguards: Shared release-approval rules apply.
Connects product, engineering, QA, research, security and platform.
Protects
- Connects product, engineering, QA, research, security and platform.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Risk-based prioritizationMIN–MAX operating system · evidence-derived · currentRisk-based prioritization. governance. Broader team: MIN–MAX operating system. Mission: Uses hard stops, bounded deferrals, immutable evidence and rollback. Operating safeguards: Shared release-approval rules apply.
Uses hard stops, bounded deferrals, immutable evidence and rollback.
Protects
- Uses hard stops, bounded deferrals, immutable evidence and rollback.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Decision-right designMIN–MAX operating system · evidence-derived · currentDecision-right design. governance. Broader team: MIN–MAX operating system. Mission: Agents gather and challenge evidence; people retain consequential authority. Operating safeguards: Shared release-approval rules apply.
Agents gather and challenge evidence; people retain consequential authority.
Protects
- Agents gather and challenge evidence; people retain consequential authority.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Research disciplineMIN–MAX operating system · evidence-derived · currentResearch discipline. governance. Broader team: MIN–MAX operating system. Mission: Keeps simulation, human evidence, operation and adjudication distinct. Operating safeguards: Shared release-approval rules apply.
Keeps simulation, human evidence, operation and adjudication distinct.
Protects
- Keeps simulation, human evidence, operation and adjudication distinct.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Accessibility and inclusionMIN–MAX operating system · evidence-derived · currentAccessibility and inclusion. governance. Broader team: MIN–MAX operating system. Mission: Treats keyboard, viewport, comprehension and lower-numeracy contexts as product evidence. Operating safeguards: Shared release-approval rules apply.
Treats keyboard, viewport, comprehension and lower-numeracy contexts as product evidence.
Protects
- Treats keyboard, viewport, comprehension and lower-numeracy contexts as product evidence.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Responsible AI leadershipMIN–MAX operating system · evidence-derived · currentResponsible AI leadership. governance. Broader team: MIN–MAX operating system. Mission: Keeps live AI reachable but disabled until exact evidence exists. Operating safeguards: Shared release-approval rules apply.
Keeps live AI reachable but disabled until exact evidence exists.
Protects
- Keeps live AI reachable but disabled until exact evidence exists.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Measurement disciplineMIN–MAX operating system · evidence-derived · currentMeasurement discipline. governance. Broader team: MIN–MAX operating system. Mission: Rejects composite passes and keeps denominators and hard gates explicit. Operating safeguards: Shared release-approval rules apply.
Rejects composite passes and keeps denominators and hard gates explicit.
Protects
- Rejects composite passes and keeps denominators and hard gates explicit.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Learning-loop designMIN–MAX operating system · evidence-derived · currentLearning-loop design. governance. Broader team: MIN–MAX operating system. Mission: Turns findings into reusable contracts, regressions and controls. Operating safeguards: Shared release-approval rules apply.
Turns findings into reusable contracts, regressions and controls.
Protects
- Turns findings into reusable contracts, regressions and controls.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Executive communicationMIN–MAX operating system · evidence-derived · currentExecutive communication. governance. Broader team: MIN–MAX operating system. Mission: States what changed, what remains blocked and what is not authorized. Operating safeguards: Shared release-approval rules apply.
States what changed, what remains blocked and what is not authorized.
Protects
- States what changed, what remains blocked and what is not authorized.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review
Scalable operating modelsMIN–MAX operating system · evidence-derived · currentScalable operating models. governance. Broader team: MIN–MAX operating system. Mission: Builds repeatable roles, manifests, scenarios and protected handoffs. Operating safeguards: Shared release-approval rules apply.
Builds repeatable roles, manifests, scenarios and protected handoffs.
Protects
- Builds repeatable roles, manifests, scenarios and protected handoffs.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Work available for review
- Work available for review
- Work available for review

Intended customer and business outcomes

Customer TrustMIN–MAX operating system · evidence-derived · currentCustomer Trust. operations. Broader team: MIN–MAX operating system. Mission: Canonical state, uncertainty, actionability and parity. Operating safeguards: Shared release-approval rules apply.
Canonical state, uncertainty, actionability and parity.
Protects
- Canonical state, uncertainty, actionability and parity.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Decision ClarityMIN–MAX operating system · evidence-derived · currentDecision Clarity. operations. Broader team: MIN–MAX operating system. Mission: A decision-first experience with one executable next action. Operating safeguards: Shared release-approval rules apply.
A decision-first experience with one executable next action.
Protects
- A decision-first experience with one executable next action.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Financial IntegrityMIN–MAX operating system · evidence-derived · currentFinancial Integrity. operations. Broader team: MIN–MAX operating system. Mission: Unknown-versus-zero and monthly-price-basis protection. Operating safeguards: Shared release-approval rules apply.
Unknown-versus-zero and monthly-price-basis protection.
Protects
- Unknown-versus-zero and monthly-price-basis protection.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Operational SafetyMIN–MAX operating system · evidence-derived · currentOperational Safety. operations. Broader team: MIN–MAX operating system. Mission: Fail-closed gates, bounded credentials and rollback. Operating safeguards: Shared release-approval rules apply.
Fail-closed gates, bounded credentials and rollback.
Protects
- Fail-closed gates, bounded credentials and rollback.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Release ConfidenceMIN–MAX operating system · evidence-derived · currentRelease Confidence. operations. Broader team: MIN–MAX operating system. Mission: Exact source, protected CI and independent review. Operating safeguards: Shared release-approval rules apply.
Exact source, protected CI and independent review.
Protects
- Exact source, protected CI and independent review.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Human UnderstandingMIN–MAX operating system · evidence-derived · futureHuman Understanding. operations. Broader team: MIN–MAX operating system. Mission: An explicit evidence layer that remains unproven. Operating safeguards: Shared release-approval rules apply.
An explicit evidence layer that remains unproven.
Protects
- An explicit evidence layer that remains unproven.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Learning VelocityMIN–MAX operating system · evidence-derived · currentLearning Velocity. operations. Broader team: MIN–MAX operating system. Mission: Incidents become reusable regressions and controls. Operating safeguards: Shared release-approval rules apply.
Incidents become reusable regressions and controls.
Protects
- Incidents become reusable regressions and controls.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
ScalabilityMIN–MAX operating system · evidence-derived · currentScalability. operations. Broader team: MIN–MAX operating system. Mission: Repeatable roles, scenarios, contracts and handoffs. Operating safeguards: Shared release-approval rules apply.
Repeatable roles, scenarios, contracts and handoffs.
Protects
- Repeatable roles, scenarios, contracts and handoffs.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Responsible AIMIN–MAX operating system · evidence-derived · currentResponsible AI. operations. Broader team: MIN–MAX operating system. Mission: AI assists evidence while deterministic logic and people retain authority. Operating safeguards: Shared release-approval rules apply.
AI assists evidence while deterministic logic and people retain authority.
Protects
- AI assists evidence while deterministic logic and people retain authority.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence
Evidence-Based GrowthMIN–MAX operating system · evidence-derived · futureEvidence-Based Growth. operations. Broader team: MIN–MAX operating system. Mission: Beta and production expand only after bounded proof. Operating safeguards: Shared release-approval rules apply.
Beta and production expand only after bounded proof.
Protects
- Beta and production expand only after bounded proof.
What it uses and produces
- Uses: Product evidence and decision context
- Produces: A decision brief or result for review
Scope and safeguards
- Act only within the named responsibility boundary
- Shared release-approval rules apply
Primary handoffs
- Passes a reviewable result to the next decision owner or independent reviewer
Handoffs and relationships
- Contributes bounded evidence
- Contributes bounded evidence
- Contributes bounded evidence

Accessible concept outline

Failure layers and programme codes, without the canvas.

Use the glossary to look up the terms in the operating model.

Product defect: The product behaves incorrectly, contradicts its intended decision model, creates a dead end, or presents an unsafe or misleading customer state. Example: A page displays “Review work-from-home requirement” even though the environment cannot execute the work-from-home question. Customer consequence: The visitor reaches a dead end or receives an untrustworthy decision. May require product remediation.
Test defect: The test expectation, fixture, locator, or assertion is incorrect even though the intended product behavior works. Example: A diagnostic expected the message field’s accessible name to equal one exact string even though the field was correctly labelled and described. Customer consequence: Delivery is blocked by faulty evidence rather than faulty product behavior. May require test remediation.
Evidence-harness defect: The mechanism used to create or retain evidence fails before it can validly judge the product. Example: An evidence package could not start because its output directory or attestation-output contract was invalid. Customer consequence: The product remains unadjudicated; no pass or failure may be inferred. Requires valid evidence before judging the product.
Continuous-integration defect: The automated integration pipeline, test runner, or build orchestration changes or destabilizes the evidence environment. Example: Automatic Continuous Integration (CI) Git-diff capture changed the checkout to a shallow repository after browser tests, invalidating later source-custody checks. Customer consequence: Release evidence cannot be trusted until the pipeline behavior is corrected. Requires pipeline or test-runtime remediation.
Environment defect: The deployed or test environment is incompatible with the approved source, schema, runtime, access, viewport, or binding contract. Example: A QA database lacked required foundational tables, so the customer journey could not be executed end to end. Customer consequence: Product behavior remains unproven in that environment. Requires environment reconciliation or a new environment.
Quality Assurance (QA): The independent validation activity or role that challenges source and intended product behavior. Quality Engineering is the broader team or function. Quality Assurance is the independent validation activity or role; neither label is a claim of extra staffed headcount.
Continuous Integration (CI): Automated source, build, test, security, and evidence checks performed when changes are integrated. CI may block release. CI cannot approve its own exception. A passing CI check does not replace human product ownership or UAT.
Agent Customer-Journey Simulation (M2A): A deterministic agent simulation that exercises frozen personas across required product paths. M2A can expose path defects and contradictions, but it cannot establish human understanding.
Agent Customer-Journey Simulation, Round 2 (M2A-R2): A deterministic second-round simulation that evaluates frozen personas across required product paths. Official M2A-R2 is not executed and not authorized. It is agent simulation, not human research or human-comprehension evidence.

Evidence-history rule: A later green run does not erase retained failed evidence. It shows that the corrected candidate passed its own exact gate.

The delivery sequence at a glance

What holds the system together

One person owns the synthesis
Specialists return useful inputs. An accountable owner resolves conflicts and assembles the result.
Add a role for a specific need
A distinct skill, context, permission or independent check can justify another agent. Name that contribution before adding the role.
Measure the deliverable
Track whether the work is complete, usable and supported by evidence. Handoffs and status updates are costs to manage.

Read Agentic Delivery Notes

Delivery model

From ambiguity to accountable delivery

The delivery model protects the decision, not the process for its own sake.

The product combines comparison logic, evidence quality and renter preferences. Bounded workflows make assumptions and handoffs visible; independent review challenges the result; the accountable product owner decides whether the evidence is sufficient to proceed.

Delivery evidence boundary

Evidence boundaries

“Built” and “proven” answer different questions.

A working prototype can demonstrate product behavior without proving that renters understand, trust or repeatedly use it. Each public claim stays attached to the evidence that supports it.

Working prototype

is not

customer validated

Internal evaluation

is not

external beta

AI-assisted analysis

is not

human judgment

Delivery independent review

Four operating principles

Clear boundaries make speed useful.

Start with the renter decision and the consequence of getting it wrong.

Separate source readiness, deployed behavior and human understanding.

Give implementation and independent review distinct responsibilities.

Keep prioritization, risk acceptance and release decisions human-owned.

Continue the story

The operating model reflects how I lead.

I organize this private build around explicit responsibilities, independent review and an accountable decision. My career history shows the enterprise product and people-leadership experience I bring to that work.

Connect the model to my experience Read the product story behind the model Discuss a role

Sources and measurement

Delivery model

From ambiguity to accountable delivery The delivery model protects the decision, not the process for its own sake. The product combines comparison logic, evidence quality and renter preferences. Bounded workflows make assumptions and handoffs visible; independent review challenges the result; the accountable product owner decides whether the evidence is sufficient to proceed.

Delivery independent review

Four operating principles Clear boundaries make speed useful. 01 Start with the renter decision and the consequence of getting it wrong. 02 Separate source readiness, deployed behavior and human understanding. 03 Give implementation and independent review distinct responsibilities. 04 Keep prioritization, risk acceptance and release decisions human-owned.

Delivery evidence boundary

A working prototype can demonstrate product behavior without proving that renters understand, trust or repeatedly use it.

Evidence boundaries “Built” and “proven” answer different questions. A working prototype can demonstrate product behavior without proving that renters understand, trust or repeatedly use it. Each public claim stays attached to the evidence that supports it. Working prototype is not customer validated Internal evaluation is not external beta AI-assisted analysis is not human judgment

A small team. A clear path to delivery.

From a product question to a release decision.

Frame

Build

Review

Decide

A renter’s next step had to work.

The next action led nowhere.

The change had to stop.

Connect the action to a working capability.

Verify the corrected path in deployment.

One mission. Bounded roles. Independent evidence.

How we classify a failure

Product defect

Test defect

Evidence-harness defect

Continuous-integration defect

Environment defect

QA, CI and agent-simulation codes

See how I organize product delivery

Seven independent evidence layers

The complete system, without the canvas.

Product mission

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Remaining blockers

Accountable product leadership

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Product-company functions

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

Handoffs and relationships

Protects

What it uses and produces

Scope and safeguards

Primary handoffs

A small team.
A clear path to delivery.