State model
A partitioned event log feeds time-windowed operational state. Per-entity sequence and watermark prevent late or duplicate events from appearing current.
Payment flows, settlement queues, collateral calls and credit lines change throughout the day across several legal entities and currencies.
A time-aware control plane reconciles current positions, projects bounded stress paths and proposes funding or collateral moves with explicit dependencies.
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.
Typed commands, state changes and receipts cross an event spine without surrendering domain ownership.
Permitted workThe design may reconcile positions and simulate options. Booking, limit consumption, valuation approval and market action remain controlled systems and roles.
Consistency ruleUse event time, valuation time, legal-entity scope and source cut-offs; label forecast, pending and settled state separately.
Hard boundaryThe model is not a system of record, identity provider, policy authority or proof that an external effect occurred.
Events or conversation turns require a decision before the underlying situation changes, with strict latency and back-pressure constraints.
The model cannot execute a movement; deterministic limits, segregation rules and dual approval govern every effect.
A deterministic outer workflow contains model-led work inside typed, observable calls. Dashed messages remain proposals until policy or a human grants authority.
A partitioned event log feeds time-windowed operational state. Per-entity sequence and watermark prevent late or duplicate events from appearing current.
Use streaming ingestion for signals, low-latency feature or state reads for the hot path, and asynchronous enrichment outside the decision budget.
Point-in-time correctness takes precedence over the newest unqualified value. Late events trigger correction or review instead of mutating the old decision invisibly.
These roles are deliberately vendor-neutral. Each can be independently owned, versioned and replaced.
Evaluates identity, purpose, capability, amount, risk tier and policy version; returns allow, deny, step-up or human-review with reasons.
Runs a versioned state machine with explicit waits, deadlines, retries, compensations, human tasks and terminal states.
Builds time-qualified projections from source events and reconciliations without becoming the legal system of record.
Presents claims beside evidence, alternatives, uncertainty, missing information, permitted actions and current custody.
Appends request, versions, policy result, model proposal, approval, action receipt, readback, correction and custody events under one correlation key.
Durable records carry provenance, authority, effect and custody without turning a transcript into an uncontrolled memory store.
Use event time, valuation time, legal-entity scope and source cut-offs; label forecast, pending and settled state separately.
The selected design is not universally superior. It is the safer fit for this boundary and failure cost.
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.
Keep consequential state transitions deterministic and use models inside bounded steps.
Let the model choose the complete path and recovery sequence.
Cost acceptedThe shell reduces flexibility, but makes deadlines, retries, permissions and recovery testable.
Use event-fed projections for scale and direct readback for consequential effects.
Fan out to all systems of record for every interaction.
Cost acceptedRead models introduce lag and reconciliation work, but reduce source load and make cross-system views feasible.
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.
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.
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.
partition_rate = peak_events_per_second / active_partitionsdecision_budget = ingest + state_read + policy + score + action_commitbacklog_clear_time = queued_events / recovery_throughputRestrict sensitive position and counterparty data to entitled desks and exclude it from reusable prompts or broad telemetry.
A design is production-ready only when teams can prove what happened, recover it and change it safely.
late market or position event test
limit breach under stale data test
valuation-source reconciliation
dual-authorisation and booking readback
Source cut-offs, reconciled positions, stress assumptions, limit checks, approvals, effect receipts and post-move readback.
Treasury operators select and authorise every funding or collateral action.
Start with advisory intervention and measured shadow scoring. Increase automation only when peak-load, late-event and fallback tests preserve the control outcome.
Treasury or desk, market risk, operations, finance and technology owners accept the design.