Projects

Hackathon agent for incident response: a GPT-5.5 supervisor calls code, company, memory and action subagents, then opens a real PR, Gmail draft and Slack post.

Research investigating LLM interpretability for financial prediction using attention analysis and causal attribution.

Co-founded medical education startup using AI-powered simulations for medical students and physicians.

Won 3rd place in Anthropic track at YC-hosted hackathon - innovative insurance protocol for AI agent transactions.

Six Claude Code plugin bundles (writing, research vault, explainer videos, NetLogo, dev tools, life) and seven multi-agent workflows, one-command install.

AI agents that automatically improve through prompt evolution and dynamic tool creation.

MCP server that enables AI-powered blogging capabilities directly within development environments through standardized protocol integration.

Chrome extension that labels each page visit on-task or distraction against your stated goal, stores sessions in Neo4j, nudges you, and writes a focus report.

CLI that extracts atomic notes from Obsidian sources with an LLM, dedupes them by title and embedding distance, and builds linked context bundles.

Universal MCP server synchronization tool with no external dependencies - pure Python standard library.