Sprint 16

• ORE import pipeline complete: dispatch pipeline, mappers for every instrument family, supporting files, and golden roundtrip tests.
• Workflow engine established with generalised event-driven FSM and party provisioning saga.
• Market data subsystem end-to-end: parser, schema, repo, service, and Qt UI.
• Qt plugin architecture introduced with 7 domain plugins split from the monolith.
• Service lifecycle controller with unified NATS heartbeat landed.
• Reporting and analytics: DQ-sourced report definitions and pricing engine.
• ~20 Windows portability fixes landed.

Dual-axis bar chart. PRs (left axis) and commits (right axis) per day. A high commits-to-PR ratio may indicate scope creep.

Lines added (green) and deleted (red) per day. Building work produces mostly additions; refactoring produces a mix. Days with no churn may indicate blockers.

Hours from PR open to merge, one bar per PR. Long bars indicate review bottlenecks. Generated only when PR data is available.

Line chart tracking stories marked DONE during the sprint. Steady upward slope is healthy; plateauing signals a stall.

• ORE import landed end-to-end inside the sprint — pipeline + mappers for all six instrument families + supporting files + server-side service.
• Workflow engine generalised mid-sprint and immediately consumed two sagas (party provisioning + ORE import); the engine becomes load-bearing infrastructure rather than party-specific machinery.
• Market data subsystem from scratch in one sprint: parser + schema
  • domain + repository + service + UI.
• Trade detail dialog unification (six merge PRs + IInstrumentForm registry) closes a long-running UX inconsistency.
• Qt plugin architecture (ores.qt.api + 7 domain plugins) prepares the codebase for v1's modularity needs.
• Windows portability sweep — 20+ fixes — gets Windows CI back to green and keeps it there.
• Asset class unification finally separates risk-taxonomy (asset_class) from structural-routing (product_type).
• Sprint hit 89 tasks across 22 stories; everything either DONE or documented as next-version work.

• 157h sprint — by far the longest. The breadth was hard to keep coherent + the review queue was substantial.
• Per-type instrument table refactor (rates / FX / equity) repeats parts of the sprint-15 CDM modelling work; that we needed both passes is evidence the first design wasn't quite right.
• Trade detail dialog merge surfaced several regressions in intermediate PRs (CDO zero values, new-trade tab visibility, etc.) that lingered across PRs before being cleaned up.
• Workflow engine generalisation came after the first concrete saga rather than before — saga-2 (ORE import) had to wait while the engine grew up.
• Windows portability fixes were largely reactive — surfaced by CI rather than prevented by deliberate cross-platform discipline.

• ores.controller owns service lifecycle; the dashboard is the visible window into it.
• ores.qt is now ores.qt.api + 7 domain plugins; the monolith is gone.
• TradeDetailDialog drives every instrument family via the IInstrumentForm registry.
• ORE import is end-to-end with mappers for every family + server- side service + supporting market data / fixings / calendar / conventions.
• Workflow engine is general-purpose; party provisioning and ORE import are both event-driven sagas on top of it.
• Market data is a real component with its own subsystem.
• asset_class and product_type are recognised as distinct concepts with shared refdata source of truth.
• Per-type instrument tables (rates / FX / equity) replace the generic instruments table from sprint 15.
• Party isolation RLS gaps closed; trading instrument subtables have proper RLS.
• UTC-everywhere timestamp policy enforced from DB to wire format.
• Windows is a first-class build target with the ~20 portability fixes that landed this sprint.

The bulk of v0's mission is now delivered: every ORE entity type is modelled, importable, and manageable through the GUI; the platform has multi-tenant + multi-party RLS, JWT-based auth, NATS-based microservices, a compute grid, a scheduler, a workflow engine, an analytics + reporting stack, and a Qt plugin architecture. v0 remains open — Sprint 17 carries the remaining work and any follow-ups surfaced here.