sheet
- role
- Solo
- period
- 2026.08 → now
- context
- Independent project. Phase 9 of 9: final benchmark in progress
- status
- ongoing
- code
- public
- stack
- Python, LangChain, Chroma, Ollama, bge-reranker, mlx-lm, LangSmith
- links
results
Finds the right passage for 83/84 exam questions (hybrid search, contextual retrieval, reranking)
Passes 70 of 84 PADI exam questions with Sonnet 5, graded by an LLM judge
Haiku 4.5 lands one question behind Sonnet 5 (69 vs 70 of 84)
skills
- technical skills
- domains
Independent project