Portfolio Chat Agent
Case study for a conversational AI chat agent built with Python, GROQ API, Qdrant Vector DB, and a RAG pipeline to answer natural language queries about Nitesh Mehta's background and projects.
PythonGROQ APIQdrant Vector DBRAG
Engineering challenge
Portfolio visitors have no structured way to explore background, experience, and projects — static pages force linear reading with no ability to ask targeted questions.
Technical solution
A RAG-backed conversational agent retrieves relevant personal context from a Qdrant vector store and feeds it to an LLM via GROQ API, grounding every answer in real data.
Measured outcome
Visitors can ask questions in plain English and receive grounded, accurate answers about experience, projects, and background — without scrolling through the entire site.