Manish Singh logo
Back to Projects

Portfolio Assistant

AI-powered assistant that answers questions about my portfolio, experience, and projects on my behalf

Project Overview

The Portfolio Assistant is an intelligent AI chatbot built with LangChain and RAG (Retrieval-Augmented Generation) technology. It serves as a virtual representative that can answer questions about my portfolio, work experience, projects, skills, and more - all on my behalf.

This project demonstrates advanced AI integration in web applications, showcasing how modern LLM (Large Language Model) technology can be used to create interactive, context-aware assistants that provide accurate and personalized responses based on portfolio data.

Key Features

  • AI-Powered Responses: Uses advanced language models to generate natural, conversational responses about my portfolio and experience
  • Context-Aware Conversations: Maintains conversation context and can answer follow-up questions based on previous interactions
  • Real-Time Communication: Instant responses via WebSocket connections for seamless, real-time chat experience
  • Smart Email Integration: Automatically detects connection intent and requests email addresses when users express interest in connecting

Technical Stack

  • Backend: Node.js with Express, Socket.IO for real-time communication, and MongoDB for message storage
  • AI & LLM: LangChain framework with RAG (Retrieval-Augmented Generation) for context-aware responses, integrated with OpenAI or similar LLM providers
  • Real-Time Communication: Socket.IO for bidirectional WebSocket communication, enabling instant message delivery
  • Frontend: Next.js React application with TypeScript, providing a modern and responsive chat interface

How It Works

  1. User asks a question about my portfolio, experience, or projects through the chat interface
  2. The system uses RAG to retrieve relevant information from my portfolio data and knowledge base
  3. LangChain processes the question with context and generates an appropriate response
  4. The response is delivered in real-time via WebSocket to the user
  5. If the user expresses interest in connecting, the assistant automatically requests their email address

Try It Out

The Portfolio Assistant is available throughout the portfolio website. Look for the chat button in the bottom-right corner to start a conversation!

Interactive Demo

The chat interface is embedded on this page. You can ask questions like 'Tell me about your experience' or 'What projects have you worked on?' and the assistant will respond on my behalf.