RAG Based AI Chatbot

Project Github

Please find the project source code : project git repository

Project Documentation

This project is a web application developed to simplify the exploration of Renesas electronics device designs through AI-powered search and conversational interaction. The application is engineered to crawl the Renesas website, extract device descriptions and corresponding block diagrams, and organize this data for intelligent retrieval.

The development process involved several key components: HTML content is parsed using BeautifulSoup to extract relevant information, which is then cleaned and transformed into document objects. These documents are embedded using Hugging Face models and stored in a FAISS vector database to support fast similarity-based search. The frontend, built with Streamlit, enables users to input queries, view the top two matching devices, and engage with a chatbot powered by OpenAI’s GPT-3.5-turbo to ask questions about the retrieved devices.

This solution provides an intuitive and interactive experience for engineers and developers, significantly enhancing their ability to discover, analyze, and understand complex device designs. By combining AI retrieval and natural language processing, the tool streamlines technical research and contributes to more efficient design and development workflows in the electronics industry.

Output Screenshot

‍