RAG Based Interactive Resume

Overview

This interactive resume uses Retrieval-Augmented Generation (RAG) to allow recruiters and visitors to "chat" with my professional background. It can answer questions about my experience, skills, and projects by retrieving relevant context from my resume and feeding it to a Large Language Model.

Architecture

1. Ingestion: Resume PDF is parsed and chunked.
2. Embedding: Chunks are embedded using a transformer model.
3. Storage: Embeddings are stored in a Pinecone vector database.
4. Retrieval: User queries are embedded and compared against stored vectors to find relevant context.
5. Generation: Context + Query are sent to Vertex AI (Gemini) to generate a response.

Key Technologies

• Frontend: Next.js, React
• Backend: Next.js API Routes / Vercel Serverless
• Vector DB: Pinecone
• LLM: Vertex AI