Software engineerNew York, NY · ET
Open to opportunities
Dependable software.
From the data up.
I turn messy data and complex workflows into tools people can rely on. 5+ years building data pipelines, developer tools, and cross-platform apps.
Data systems Developer tools AI workflows
Selected workA closer look at the tools, systems, and interfaces I've built.
An Electron desktop app that brings clinical data validation, PDF review, and study workflows into one place. Built for the people checking records before anyone downstream trusts them.
A working LangGraph agent that plans, calls tools, and carries state across multi-step tasks. A walkthrough of the orchestration behind the answers.
A React Native demo connecting authentication, realtime updates, and Postgres search to a mobile interface, with Expo and Supabase underneath.
My work has mostly been in regulated, high-stakes environments, where one wrong record is a real problem for a real person. That shapes what I build: pipelines that move privacy-sensitive data, validation that catches a bad record before anything downstream trusts it, and interactive visualizations that make the result legible.
Not all of it has been infrastructure. I've built internal tools other engineers adopted, co-founded a startup, and shipped consumer products — and I'm comfortable owning a system from the architecture call through the release. I'd rather ship something correct and boring than clever and fragile.
Correct and boring beats clever and fragile.
The languages and platforms I use to take a system from architecture to release.
AI agents, local models, and developer tooling — explained through working systems.
An OpenAI Agent Escaped Its Sandbox—and AI Had Its Wildest WeekVideo
A benchmark agent found a zero-day in its own package proxy, reached the internet, and pulled the answers from Hugging Face. Long-horizon autonomy moves the unit of safety from the single tool call to the whole trajectory.
How I Built a Local AI Talking Head That Keeps the Real PersonVideo
Instead of regenerating the person, the pipeline keeps the real recording and synthesizes only the pixels inside a mouth mask. SyncNet holds the normal track at a zero-frame offset, and a deliberate 400 ms shift moves it by 10 frames — the evaluator is tracking timing, not faces.
Why AI Agents Fail After the Demo—and What CARS24 Gets RightVideo
CARS24 reports more than a million agent conversation minutes a month while most agent demos collapse the moment the task leaves a clean prompt. The difference is a five-part contract around the model: accepted outcome, bounded permissions, evidence-based routing, trajectory evaluation, and a fallback.
/ Let's build
Interested in software engineering roles focused on data integration, developer tooling, or cross-platform product work — especially where data accuracy and reliability are non-negotiable.