Day 19: app.py - From Notebook to src/, and the Index That Went StaleI added the RAG paper to my RAG pipeline, and it told me it couldn't find anything about RAG.8 min read·parthkg.hashnode.dev11MSep 21
Day 18: pdf_loader.ipynb — Adding Groq, and Teaching RAG to Say NoMy RAG pipeline answered a question none of my PDFs could. That turned out to be a bug.7 min read·parthkg.hashnode.dev12AMSep 20
Day 17: pdf_loader.ipynb — ChromaDB, a Retriever, and a Lying ScoreMy retriever said nothing matched. The right chunk was in the database the whole time.5 min read·parthkg.hashnode.dev12AMSep 19
Day 16: pdf_loader.ipynb — 64 Pages, 359 Chunks, 384 DimensionsMy chunks were 1,000 characters long. The embedding model only reads the first 256 tokens.5 min read·parthkg.hashnode.devSep 17
Day 15: document.ipynb — What RAG Fixes and LangChain's DocumentFine-tuning makes the model memorize the textbook. RAG lets it bring the textbook into the exam.5 min read·parthkg.hashnode.devSep 16
