AI agent that writes Python code, auto-generates test cases, executes, captures errors, and self-corrects in a loop until all tests pass. Built with LangGraph + Groq Llama-3.3-70B + Streamlit.

An AI coding assistant built with Streamlit and a modular Python backend. It leverages Ollama and LangChain for private, on-device code generation, debugging, and complexity analysis across multiple languages. The project showcases skills in system architecture and building practical, self-contained developer tools.