golangLAKEHOUSE

profit/golangLAKEHOUSE

Fork 0

Commit Graph

Author	SHA1	Message	Date
root	09299a27b7	scrum 2026-05-02: materializer+replay+vectord — ship-with-fixes Cross-lineage review of 89ca72d (Opus + Kimi + Qwen3-coder). Convergent findings (≥2 reviewers): NONE. - Kimi BLOCK (materializer main.go exits 0 on validation fail): confabulation. Code does os.Exit(1) at lines 65-66. - Qwen BLOCK (saveTask race condition): confabulation. All access to inflight/pending is under s.mu. - Qwen WARN (saveAfter nil deref): confabulation. Explicit `if h.persist == nil { return }` guard at line 184. - Opus BLOCK (TestSaveTask_Coalesces): self-withdrawn in same response. Opus WARNs actioned: - Detached docstring on TestAdd_SmallIndex_ConcurrentDistinctIDs — attached. - isoDatePartition fallback comment — clarified as defense-in-depth (MaterializeAll guards upstream; branch unreachable through public surface). Disposition + verdicts in reports/scrum/_evidence/2026-05-02/. Pattern matches feedback_cross_lineage_review.md: only Opus emits BLOCK-class findings worth verifying; Kimi/Qwen single-reviewer BLOCKs failed trace verification. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-05-02 03:35:12 -05:00
root	89ca72d471	materializer + replay ports + vectord substrate fix verified at scale Two threads landing together — the doc edits interleave so they ship in a single commit. 1. vectord substrate fix verified at original scale (closes the 2026-05-01 thread). Re-ran multitier 5min @ conc=50: 132,211 scenarios at 438/sec, 6/6 classes at 0% failure (was 4/6 pre-fix). Throughput dropped 1,115 → 438/sec because previously-broken scenarios now do real HNSW Add work — honest cost of correctness. The fix (i.vectors side-store + safeGraphAdd recover wrappers + smallIndexRebuildThreshold=32 + saveTask coalescing) holds at the footprint that originally surfaced the bug. 2. Materializer port — internal/materializer + cmd/materializer + scripts/materializer_smoke.sh. Ports scripts/distillation/transforms.ts (12 transforms) + build_evidence_index.ts (idempotency, day-partition, receipt). On-wire JSON shape matches TS so Bun and Go runs are interchangeable. 14 tests green. 3. Replay port — internal/replay + cmd/replay + scripts/replay_smoke.sh. Ports scripts/distillation/replay.ts (retrieve → bundle → /v1/chat → validate → log). Closes audit-FULL phase 7 live invocation on the Go side. Both runtimes append to the same data/_kb/replay_runs.jsonl (schema=replay_run.v1). 14 tests green. Side effect on internal/distillation/types.go: EvidenceRecord gained prompt_tokens, completion_tokens, and metadata fields to mirror the TS shape the materializer transforms produce. STATE_OF_PLAY refreshed to 2026-05-02; ARCHITECTURE_COMPARISON decisions tracker moves the materializer + replay items from _open_ to DONE and adds the substrate-fix scale verification row. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>	2026-05-02 03:31:02 -05:00

Author

SHA1

Message

Date

root

09299a27b7

scrum 2026-05-02: materializer+replay+vectord — ship-with-fixes

Cross-lineage review of 89ca72d (Opus + Kimi + Qwen3-coder).

Convergent findings (≥2 reviewers): NONE.

- Kimi BLOCK (materializer main.go exits 0 on validation fail):
  confabulation. Code does os.Exit(1) at lines 65-66.
- Qwen BLOCK (saveTask race condition): confabulation. All access
  to inflight/pending is under s.mu.
- Qwen WARN (saveAfter nil deref): confabulation. Explicit
  `if h.persist == nil { return }` guard at line 184.
- Opus BLOCK (TestSaveTask_Coalesces): self-withdrawn in same
  response.

Opus WARNs actioned:
- Detached docstring on TestAdd_SmallIndex_ConcurrentDistinctIDs —
  attached.
- isoDatePartition fallback comment — clarified as defense-in-depth
  (MaterializeAll guards upstream; branch unreachable through public
  surface).

Disposition + verdicts in reports/scrum/_evidence/2026-05-02/.

Pattern matches feedback_cross_lineage_review.md: only Opus emits
BLOCK-class findings worth verifying; Kimi/Qwen single-reviewer
BLOCKs failed trace verification.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-02 03:35:12 -05:00

root

89ca72d471

materializer + replay ports + vectord substrate fix verified at scale

Two threads landing together — the doc edits interleave so they ship
in a single commit.

1. **vectord substrate fix verified at original scale** (closes the
   2026-05-01 thread). Re-ran multitier 5min @ conc=50: 132,211
   scenarios at 438/sec, 6/6 classes at 0% failure (was 4/6 pre-fix).
   Throughput dropped 1,115 → 438/sec because previously-broken
   scenarios now do real HNSW Add work — honest cost of correctness.
   The fix (i.vectors side-store + safeGraphAdd recover wrappers +
   smallIndexRebuildThreshold=32 + saveTask coalescing) holds at the
   footprint that originally surfaced the bug.

2. **Materializer port** — internal/materializer + cmd/materializer +
   scripts/materializer_smoke.sh. Ports scripts/distillation/transforms.ts
   (12 transforms) + build_evidence_index.ts (idempotency, day-partition,
   receipt). On-wire JSON shape matches TS so Bun and Go runs are
   interchangeable. 14 tests green.

3. **Replay port** — internal/replay + cmd/replay +
   scripts/replay_smoke.sh. Ports scripts/distillation/replay.ts
   (retrieve → bundle → /v1/chat → validate → log). Closes audit-FULL
   phase 7 live invocation on the Go side. Both runtimes append to the
   same data/_kb/replay_runs.jsonl (schema=replay_run.v1). 14 tests green.

Side effect on internal/distillation/types.go: EvidenceRecord gained
prompt_tokens, completion_tokens, and metadata fields to mirror the TS
shape the materializer transforms produce.

STATE_OF_PLAY refreshed to 2026-05-02; ARCHITECTURE_COMPARISON decisions
tracker moves the materializer + replay items from _open_ to DONE and
adds the substrate-fix scale verification row.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-05-02 03:31:02 -05:00

2 Commits