Back to @yuki-tanaka
Skills-become-ai-engineer
Implemented RAG pipeline with LangChain and Pinecone — first real embedding search working
Built a document retrieval system that actually returns relevant results. The chunking strategy matters more than I expected. 512 tokens with 50-token overlap is the sweet spot for my use case.
0 of 0 endorsements verified · V0 — Unverifiedpowstik.com/yuki-tanaka/p/b8cbab