Engineering has moved from writing code by hand to directing it. In skilled hands, AI is a force multiplier. In unskilled hands, it ships confident, plausible-looking bugs at scale.
AI doesn't replace engineering judgment. It amplifies whatever judgment is already there — good or bad.
Years as a U.S. National Chess Master trained one skill above all: holding many lines in your head at once and choosing well under pressure. That is exactly what multi-agent development demands.
Scope the work, give agents the right context, and keep the architecture in one head — so many streams still add up to one coherent system.
Prompt validation, automated unit tests and architectural sanity checks on every LLM output. Nothing merges on trust.
Live system state in, schema-constrained output, validation gates before anything executes — agents that act on what's real, not what they assume.
Leading department-wide moves from manual coding loops to AI-native pipelines, with shared validation standards.
Using AI to deliver more, faster — without trading away quality. Every AI-built solution goes through the same rigor as hand-written code: clear intent, real tests, and architectural review before anything ships. The result is more output and fewer defects, not one at the expense of the other.
Python frameworks for RenderDoc GPU profiling and Perfetto-based real-time graphics tests, run on physical Oculus headsets — measuring real devices under real load.
A C#/SQL automation engine driving 10,000+ data-driven tests across an enterprise suite, plus deployment orchestration with Cypress and custom Slackbots.
React/Jest end-to-end suites for core messaging infrastructure. Earned Meta's internal “Jest-Hero” award.
NLP and transformer models combined with prompt engineering to route support tickets automatically.
From set-top boxes to VR headsets to AI pipelines — always where software meets real systems, and where reliability isn't optional.
AI orchestration, LLM-driven automation architecture, and leading the department's move to AI-native pipelines.
GPU profiling and graphics performance frameworks for VR headsets; Messenger end-to-end automation.
A 10,000+ test data-driven C#/SQL automation engine, plus deployment orchestration via Cypress and custom Slackbots.
Company-wide API test suites in Python, penetration testing, and distributed teams validating VOD on set-top box hardware.
“Ron is a highly intelligent, motivated, and creative problem solver… I have no hesitation in recommending him to anyone.”
“…contributed greatly to the quality and supportability of our highly complex enterprise application suite.”
“…he excelled in writing automation and test tools to reduce testing time and increase coverage.”
L.A. County Champion and undefeated USAT National Champion.
Travels for the history — most recently, Athens.
Open to AI-native engineering roles and orchestration work. Based in Los Angeles.