Projects

ZoomMCP is a Model Context Protocol (MCP) server that gives AI assistants like Claude direct control over Zoom — creating meetings, managing participants, pulling recordings, and more, all through natural language. No more switching tabs or hunting through menus; you describe what you want and the AI handles it.

MCP is an open standard that lets AI models talk to external tools through a structured protocol. ZoomMCP implements that protocol for the Zoom API, acting as the bridge between your AI assistant and your Zoom account. It runs locally, keeps your credentials private, and exposes Zoom's capabilities as callable tools the model can reason about and chain together.

The project was built to solve a real friction point at the intersection of enterprise communication and the AI agent ecosystem — and to prove that serious API integrations don't require complex infrastructure to feel magical.

Cultural events — concerts, gallery openings, street fairs, community nights — happen constantly, get shared briefly on social media, and then disappear. Culture Hub is a platform built to stop that from happening. It surfaces events, aggregates them in one place, and keeps a record so nothing slips through the cracks.

The problem it solves is a familiar one: you find out about something great three days after it happened, or you meant to go but couldn't find the details again. Culture Hub treats local culture as data worth preserving — not just a feed to scroll past.

Built with a scraping and aggregation pipeline that pulls events from multiple sources, normalizes them, and presents them in a browsable, searchable format. The goal is to make your city's cultural life feel less ephemeral and more like something you can actually participate in.

Your utility company has been collecting detailed data about your energy usage for years. Most people never look at it. This project is about looking at it — and finding out what it actually tells you.

Electricity Data pulls in granular hourly consumption data from Con Edison, cleans and structures it, and produces a set of visualizations that reveal patterns invisible in a monthly bill: when does your household peak? How does usage shift across seasons? What does a cold snap actually cost? The data was already yours — this project just makes it legible.

Built as a reproducible Python pipeline with Jupyter notebooks for exploration and a clean output layer for sharing. The focus is on honest, readable charts that communicate real insight without requiring the reader to be a data analyst.