Safety & Systems Engineer · San Francisco, CA
// Senior Software Systems Engineer · Latitude AI · Founder, Control Action
A decade building safe, reliable cyber-physical systems at the frontier of autonomous vehicles — from powertrain calibration at Ford to L4 robotaxi deployment at Cruise to perception feature architecture at Latitude AI. Specialized in SOTIF, STPA, and ML validation for systems that have to work when it matters.
Applying STAMP-based hazard analysis to autonomous and ML-driven systems. Authoring FMEAs, conducting STPA, and writing SOTIF documentation that actually drives design decisions — not just compliance checkboxes.
Designing verification and validation programs for ML-driven autonomous capabilities — bounding scenario definition, structured track and fleet test design, statistical test planning, and sim-to-real gap analysis.
Authoring concept of operations, defining system I/Os, writing feature and functional requirements, and designing verification plans for perception and autonomy features across the full development lifecycle.
A decade of embedded experience across Ford, Cruise, and Latitude AI — from powertrain calibration to L4 robotaxi safety cases to perception system architecture for ADAS features on production vehicles.
Control Action is an independent consulting practice applying SOTIF and STPA frameworks to AI and embodied systems — robotics, autonomous vehicles, drones, and any ML system operating in the physical world.
The practice was built on a simple premise: most ML validation programs are designed for software, not for systems that crash. The frameworks that work for AV safety cases — STAMP, STPA, SOTIF — are underutilized everywhere else. That's the gap.
Writing on the Substack applies these lenses to broader AI and tech — accessible analysis for practitioners who want to think more rigorously about the systems they build.
Read Control Action on Substack ↗Hazard analysis, safety case review, and SOTIF documentation for ML-driven features in autonomous platforms.
Building verification and validation plans for new autonomous capabilities — bounding scenarios, test design, metrics, and deployment gates.
Reviewing functional architectures, requirements, and system designs against safety standards and deployment constraints.
Embedded support for early-stage robotics and AV teams building their first safety-engineering infrastructure.
Open to consulting engagements for AV, robotics, and embodied AI teams — particularly early-stage companies building their first serious safety engineering practice.