///AI/SW
Active
RAG-006

Rag-Pipeline

RAG-powered codebase API

Ask your codebase questions in plain English and get instant answers with exact file locations. Turn hours of digging through unfamiliar code into simple conversations.

///Case Study
Challenge
01

Generic AI answers don't know your codebase. Wanted a chatbot that could answer questions about my own repositories with real context.

Sec.01
Approach
02

RAG pipeline built with .NET 9 and Semantic Kernel. Indexes GitHub repos, chunks code into embeddings, stores vectors locally, and retrieves relevant context for Claude to generate answers.

Sec.02
Outcome
03

Runs locally on-demand. Not a daily tool, but invaluable when you need to ask "where does this project handle X?" and get a real answer.

Sec.03