
AI-powered Chrome extension that discovers job opportunities hidden in LinkedIn posts. Searches posts via the LinkedIn API, filters noise with AI, extracts structured job data, and monitors for new opportunities in the background.
Conversational AI agent for structured symptom documentation. Built during the Berkeley AgentX Hackathon. Guides users through a schema-driven dialogue, supports multiple symptom threads with persistence.
Model Context Protocol (MCP) server for automating LinkedIn post searching and tracking via your AI assistant (Cursor/Claude). Uses Playwright for authenticated scraping, stores everything locally in SQLite, and includes a React dashboard for filtering and application tracking.
On-device Chrome extension that transforms YouTube videos into interactive learning. Built for the Google Chrome Built-in AI Challenge. Chat with videos and generate quizzes privately using Gemini Nano, with a lightweight RAG pipeline (local embeddings + vector search) for long transcripts.