State model
Append-only case events feed a current-state projection. Evidence rows keep source, observation time, validity, correction history and narrative usage.
A card dispute mixes scheme rules, merchant evidence, customer testimony and transaction telemetry with strict response deadlines.
The assembler normalises artefacts, maps them to effective rules, tests timeline and amount consistency, and drafts a cited case position with explicit evidence gaps.
Organisations, systems and operating conditions are intentionally anonymised and recomposed. The design demonstrates engineering and banking-domain reasoning; it does not represent a named client estate, vendor product or measured production result.
A shared authority core coordinates independently owned capability cells, domain systems and operating evidence.
Permitted workThe AI route may explain and prepare an instruction. Payment authentication, limits, screening, posting and settlement remain deterministic control points.
Consistency ruleUse idempotent commands, monotonic status and independent ledger or rail readback; timeout never means not executed.
Hard boundaryThe model is not a system of record, identity provider, policy authority or proof that an external effect occurred.
An alert or case requires evidence fan-out, deterministic triage, residual model reasoning and a defensible human disposition.
Only deterministic rule checks can assert eligibility; model output is limited to extraction, comparison and question generation.
A deterministic outer workflow contains model-led work inside typed, observable calls. Dashed messages remain proposals until policy or a human grants authority.
Append-only case events feed a current-state projection. Evidence rows keep source, observation time, validity, correction history and narrative usage.
Use asynchronous ingestion and bounded parallel reads. Slow providers have independent timeout and circuit-breaker policy so one source cannot exhaust the case tier.
Pin policy and evidence versions at case open. Later corrections append and may trigger reassessment; they do not rewrite the historic basis silently.
These roles are deliberately vendor-neutral. Each can be independently owned, versioned and replaced.
Accepts a typed evidence request and returns the minimum permitted facts with source, event time, observation time, validity and exclusions.
Appends request, versions, policy result, model proposal, approval, action receipt, readback, correction and custody events under one correlation key.
Serves owned, audience-qualified and effective-dated content; exposes supersession, withdrawal and dependency metadata.
Presents claims beside evidence, alternatives, uncertainty, missing information, permitted actions and current custody.
Evaluates identity, purpose, capability, amount, risk tier and policy version; returns allow, deny, step-up or human-review with reasons.
Durable records carry provenance, authority, effect and custody without turning a transcript into an uncontrolled memory store.
Use idempotent commands, monotonic status and independent ledger or rail readback; timeout never means not executed.
The selected design is not universally superior. It is the safer fit for this boundary and failure cost.
Prefetch stable, purpose-safe facts; acquire volatile facts on demand against a time-qualified snapshot.
Load a broad customer profile at session start.
Cost acceptedThe selected design adds source calls and latency, but reduces stale data, excess exposure and accidental reuse.
Use append-only events plus a rebuildable current-state projection.
Overwrite the case row with its latest status.
Cost acceptedReplay and storage are more complex, but point-in-time reconstruction and correction lineage remain possible.
Federate authoring while centralising lifecycle metadata, validation and serving rules.
Create one centrally authored knowledge corpus.
Cost acceptedFederation requires stronger contracts and owner discipline, but preserves domain accountability and release velocity.
Keep people at irreversible, ambiguous and policy-exception points; sample lower-risk automated outcomes independently.
Require the same manual approval at every step.
Cost acceptedRisk-tiering reduces review load but needs calibrated thresholds, sampling and immediate withdrawal of authority when drift appears.
Compile stable decision logic and retain retrieval for explanation and residual ambiguity.
Ask a model to interpret the source document for every request.
Cost acceptedRule compilation needs controlled change, but creates repeatable decisions, regression tests and clear exceptions.
Retries are bounded by knowledge of business effect; unknown outcome remains visible, owned and independently reconciled.
Actual thresholds belong to accountable service owners. The design exposes the equations and observables that those owners must baseline.
peak_enrichment_qps = peak_case_rate x sources_per_casereview_hours = residual_cases / investigator_throughputevidence_storage = cases x evidence_rows x retention_windowRestrict payment detail to the authenticated purpose and redact unnecessary counterparty data from model context and telemetry.
A design is production-ready only when teams can prove what happened, recover it and change it safely.
duplicate and delayed acknowledgement test
status change before commit test
limit and step-up authentication test
unknown-outcome recovery rehearsal
Rule version, artefact provenance, timeline, arithmetic checks, gap list and reviewer disposition.
A disputes specialist submits, contests or closes the case.
Use shadow decisions and blind samples before allowing deterministic closes. Keep agent-touched cases under human disposition until operating evidence supports narrower review.
Payments product, fraud, financial-crime, operations and service-resilience owners approve the route.