Commit Graph

3 Commits

Author SHA1 Message Date
Tobi Lutke
431f6e505b
Fix qmd embed crash and resolve all TypeScript errors
- Fix ReferenceError in vectorIndex(): firstResult was used but never
  defined. Added code to embed first chunk to get embedding dimensions.

- Fix 87 TypeScript errors across codebase:
  - formatter.ts: Define MultiGetFile type locally (was missing from store.ts)
  - collections.ts: Add non-null assertion for array access
  - mcp.ts: Fix StatusResult type to match store.ts CollectionInfo,
    add list parameter to ResourceTemplate, fix undefined checks
  - qmd.ts: Fix boolean/string type coercions, undefined array access
  - llm.test.ts: Update expandQuery tests for Queryable[] return type,
    fix array access assertions
  - store.test.ts: Add non-null assertions for array access in tests
  - eval-harness.ts: Fix array access assertion
2025-12-31 13:32:30 -04:00
Tobi Lutke
945d4b4572
Add 6 synthetic evaluation documents
Topics covered:
- API design principles
- Startup fundraising memo
- Distributed systems overview
- Product launch retrospective
- Machine learning primer
- Remote work policy

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-21 13:10:35 -04:00
Tobi Lutke
7828566333
Add evaluation harness with synthetic test documents
- 6 public-style documents covering diverse topics
- 18 test queries: 6 easy, 6 medium, 6 hard
- Easy: exact keyword matches
- Medium: semantic/conceptual queries
- Hard: partial recall, indirect references
- Measures Hit@1, Hit@3, Hit@5 by difficulty
- Tests both search (BM25) and query (hybrid) modes

Run: bun test/eval-harness.ts

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
2025-12-21 13:10:24 -04:00