My portfolio website built with React, TypeScript, and Vite, with a live GitHub integration and MattBot, an interactive AI resume assistant built with the Gemini API.
- MattBot (AI Assistant): An interactive chatbot grounded in my professional resume data using RAG (Retrieval-Augmented Generation) principles. It can answer specific questions about my experience, education, and skills.
- Live GitHub Integration: Dynamically fetches and displays my latest projects directly from the GitHub REST API.
- CI/CD: Automated deployment pipeline to GitHub Pages via GitHub Actions.
- Frontend: React, TypeScript, Tailwind CSS
- AI/ML: Google Gemini API
- Build Tool: Vite
- Deployment: GitHub Actions & GitHub Pages
MattBot uses a Contextual Grounding approach. Resume data is stored as a structured string within the application. When a user sends a message:
- The application retrieves the latest repository data from GitHub.
- It combines the static resume data and live GitHub context into a System Instruction.
- The Gemini model processes the user's query within this specific context, ensuring answers are accurate and relevant to my educational and professional background.
