Senior Machine Learning Engineer | San Francisco, CA
Sr. Machine Learning Engineer
San Francisco, CA
Sounds cliché, but making software do something in the real world is what excites me. I specialize in zero-to-one problem solving, building prototypes that evolve into scalable systems, and shipping robotics and ML products that operate reliably in the field. Experience spans robotic arms, drones, autonomous systems, and embedded ML at scale.
San Francisco, CA | 2026 - Present
Python, TypeScript (React), Go, C# (.NET)
Autonomous Robotic Laboratory & AI Agents
Lead the AI scientist pod at an AI-run robotic lab startup; DRI for the AI experimentalist on a major pharma customer engagement; co-designed the lab's three-tier software architecture.
Built Medra OS: fully unattended bare-metal provisioning that scaled the lab from 10 to 100+ robotic workstations; absorbed an overnight pivot to consumer mini PCs, standing up 60 decks in 4 days.
Integrated lab instruments (thermocycler, qPCR, plate-handoff rail electronics and firmware) in days each, then packaged the process as an agent skill teammates used to bring up more instruments unaided.
Created a company-wide autonomous SWE agent used by 50+ engineers; most of its work is self-initiated (PR triage, fleet-alarm first response, bug-report dedup) and most automations on it were built by others.
Own the monorepo developer environment (4 languages) and an internal apps hub hosting 11 apps in regular use across the company.
San Francisco, CA | 2018 - 2026
Python, Golang, Typescript (React, Redux, React Native), Android (Java, Kotlin)
Edge ML Infrastructure
Led end-to-end design and deployment of embedded video ML pipelines powering Samsara’s AI dashcams (2M+ devices in the field).
Shipped safety-critical features (e.g. tailgating, rolling stop, lane departure) running on resource-constrained hardware.
Built QA device farms, shadow testing systems, and firmware infrastructure for rapid iteration of on-device ML.
Spearheaded offline evaluation pipelines with multimodal inputs (video, IMU, GPS, hardware-in-the-loop simulation / emulation).
Designed debugging and telemetry tools to improve observability across firmware and cloud systems.
San Francisco, CA | 2014 - 2018
Javascript (React, Redux, Leaflet, React Native), Python, Swift, PHP, Golang
Drone Platforms & Aerial Robotics
Tokyo, Japan | 2012 - 2014
Python (django, OpenRAVE), Javascript (Backbone/Marionette, WebGL)
Industrial Robotics
B.S. Computer and Systems Engineering | 2008 - 2012
Rensselaer Center for Open Source, Red & White Alumni Organization
Capstone - simulated intra-wing robot for aircraft inspection. (project link)