Build a RAG Chatbot Using DeepSeek and LlamaIndex

In this project, we’ll build an interactive AI chat agent that uses retrieval-augmented generation (RAG) to answer user questions by fetching real-time content from Wikipedia. We’ll use Chainlit to create a chat interface and settings menu, allowing users to select a Groq-hosted LLM model (like LLaMA 3 or DeepSeek) and specify a topic to index from Wikipedia. The backend integrates the Groq API for low-latency inference and LlamaIndex for document chunking, indexing, and query-based retrieval.

Learners are guided through importing required libraries, configuring user settings, indexing Wikipedia pages, and creating a ReAct-based agent that can think step by step, use tools, and return thoughtful final answers. This project offers a hands-on way to explore LLM-powered chat apps, RAG pipelines, and agentic reasoning using real-world data and modern inference stacks.