I'm a EECS + Business student at UC Berkeley MET. I build AI products, developer tools, and interfaces that make complicated systems easier to use.
Right now I'm interested in persistent context for AI, voice as an interface, and RL rooms for collaborative AI.
- Cortex — a local-first memory + taste layer that gives AI tools persistent context
- Calendera — a Cal Hacks '25 award-winning unified inbox that uses AI to turn messages across apps into events, tasks, and priorities
- DuckTape — a voice-first AI coding assistant that remembers your project across the web and terminal
- LifeOS — an autonomous health agent that turns daily signals into actions, measures the result, and remembers what worked
- RunAnywhere Pipeline Profiler — cross-platform tracing for STT → LLM → TTS pipelines, with a Perfetto-style timeline viewer
- Tech Events Tracker — a serverless site that finds local hackathons and startup events and republishes itself every day

