Hi, I'm
I build applied AI/ML systems and the evaluation frameworks that prove they work.
Recent Statistics & Data Science + Technology Management graduate from UC Santa Barbara.
I. About
I'm a recent Statistics & Data Science graduate from UC Santa Barbara, with a Technology Management Certification. I'm drawn to transforming complex systems into interpretable insights, revealing patterns that inform meaningful decisions.
Right now I'm building the real-time voice and vision stack for an AI agent that joins Discord calls and plays multiplayer games alongside you, with sub-420ms voice latency and 60+ FPS live game-state detection. I spent the first half of 2026 as a data science intern at BlueAlpha, building synthetic data generators and benchmarking workflows for Marketing Mix Modeling — work I carried into Juno, an agentic copilot that interprets MMM outputs and ships with a benchmark measuring how far its advice can be trusted. Before that I spent nine months at NCEAS designing reproducible pipelines for biodiversity data. On the side, I build projects like Axis, a portfolio analytics app that backtests its own allocation and forecasting methods, and Scalpal, a VR surgical trainer where I built the realtime backend that lets the headset, browser, and AI coach share one operating room.
Outside of code, you will find me reading books, going on runs, exploring nature, or spending time with family and friends.
II. Experience
San Francisco, CA (Remote)
Jun 2026 — Present
San Francisco, CA
Jan 2026 — Jun 2026
Santa Barbara, CA
Jul 2025 — Mar 2026
III. Projects
A handful of products and experiments. Click any project for the longer story.
Featured
Agentic copilot for Marketing Mix Model outputs — and a benchmark that measures whether its advice holds up.
A multi-agent system that turns MMM coefficients into prioritized, citation-grounded recommendations — scored against 100 ground-truth scenarios by an LLM judge at 0.897 composite and 0.875 ranking accuracy.
Demo sleeps when idle — first load may take ~50s to wake.
Portfolio analytics that tests its own methods — and reports when they don't beat the simple baseline.
An R Shiny app on live market data with QP optimizers and Ledoit-Wolf shrinkage. A walk-forward backtest of 5 allocation strategies from 2007 to 2026 found none beat equal weight on Sharpe beyond noise, while min variance cut worst drawdown from 37.9% to 20.2%.
VR surgical trainer where the operating room lives in the database — headset, browser, and AI coach all play in the same one.
A Meta Quest 3S simulator for practicing surgery with a voiced AI attending. I built the realtime backend for the four-person team: a 25-table SpacetimeDB module with 52 role-checked reducers and 23 identity-scoped views, giving every client one authorization model.
★ MHacks 2026 — Best Use of SpacetimeDB
The full experience runs on a Meta Quest 3S.
Real-time computer-vision coach for weightlifting form.
Real-time pose tracking at ~30 FPS with ~200ms feedback latency, paired with a 4-loop agentic AI system delivering personalized voice cues mid-set across 3 compound lifts.
★ BroncoHacks 2026 — Best Use of Backboard
AI guardrails against manipulative social-media content, in real time.
Distributed 5-agent architecture monitoring Instagram in real time, classifying manipulative engagement patterns via an LLM pipeline and intervening before they reach you.
★ BeachHacks 9.0 — Best Mental Health + Best Use of Fetch.ai
IV. Connect
Best way to reach me is over email. I'm also on GitHub and LinkedIn.