AI Portfolio Chatbot

An interactive portfolio website powered by RAG (Retrieval-Augmented Generation) that allows visitors to chat with an AI version of me to learn about my experience, skills, and projects.

🌐 Live Demo: anshkakkar.dev

🚀 How It Works

This project uses a RAG (Retrieval-Augmented Generation) architecture:

1. Embedding Creation: Resume content is converted into vector embeddings using OpenAI's text-embedding-ada-002 model
2. Vector Storage: Embeddings are stored in Supabase's pgvector database for efficient similarity search
3. Semantic Search: When a user asks a question, it's converted to an embedding and matched against the stored vectors
4. Context Retrieval: The most relevant resume information is retrieved based on semantic similarity
5. AI Response: OpenAI's GPT-4 generates natural, conversational responses using the retrieved context

This approach ensures accurate, contextual responses while preventing hallucinations by grounding answers in actual resume content.

🛠️ Tech Stack

• Frontend: Next.js 15, React
• Styling: Custom CSS with modern gradients and animations
• AI/ML: OpenAI API (GPT-4 & Embeddings)
• Database: Supabase (PostgreSQL + pgvector)
• Deployment: Vercel