root 390ebf0c36 IVF_PQ recall tuned from 0.80 → 0.97 via parameter sweep ...

Systematic sweep of 8 IVF_PQ configs on 100K × 768d resumes.
num_sub_vectors is the dominant lever: 48 → 192 pushes recall
from 0.795 → 0.970. Winner: partitions=500, bits=8, subs=192.
Build 61s (vs 18s baseline), acceptable for background builds.

Hybrid status: HNSW recall=1.00 at <1ms, Lance IVF_PQ recall=0.97
at 60ms. Both backends production-grade.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-04-16 22:08:34 -05:00

crates

S3 backend for Lance — hybrid operates on real MinIO object storage

2026-04-16 21:09:42 -05:00

data

Stress test suite: 9/9 passed — architecture validated

2026-03-27 22:13:27 -05:00

docs

Update PRD + PHASES.md — reflect 8-commit 2026-04-17 push

2026-04-16 20:54:05 -05:00

inbox/processed

Scheduled ingest: file watcher auto-ingests from ./inbox

2026-03-27 20:04:40 -05:00

proto

Phase 5: hardening — gRPC, observability, auth, config

2026-03-27 06:37:07 -05:00

scripts

IVF_PQ recall tuned from 0.80 → 0.97 via parameter sweep

2026-04-16 22:08:34 -05:00

sidecar

Phases 16.2 + L2 + 17 VRAM gate + MySQL + 18 Lance hybrid milestone

2026-04-16 20:24:46 -05:00

tests

Phase 2: DataFusion query engine over Parquet

2026-03-27 05:48:20 -05:00

.gitignore

Add __pycache__ to gitignore

2026-03-27 05:54:09 -05:00

Cargo.lock

Phases 16.2 + L2 + 17 VRAM gate + MySQL + 18 Lance hybrid milestone

2026-04-16 20:24:46 -05:00

Cargo.toml

Phases 16.2 + L2 + 17 VRAM gate + MySQL + 18 Lance hybrid milestone

2026-04-16 20:24:46 -05:00

justfile

Phase 4: Dioxus frontend with dataset browser and SQL query editor

2026-03-27 06:24:15 -05:00

lakehouse.toml

S3 backend for Lance — hybrid operates on real MinIO object storage

2026-04-16 21:09:42 -05:00

Description

Rust-first object storage system

6.3 GiB

Languages

TypeScript 38.4%

Rust 35.8%

HTML 13.9%

Python 7.8%

Shell 2.1%

Other 2%