AI Architect & Enterprise AI Transformation Leader
I've spent a decade at the intersection of software engineering, technical leadership, and applied AI — moving from writing full-stack code and teaching engineers, to running analytics and operations at scale, to designing and deploying production AI across an enterprise. The throughline is the same: build the system, then build the organization that can use it.
Principles that shape what I build
Systems, not demos
A demo proves something can work once. A system proves it works repeatedly, measurably, and safely. I build for the second — with evaluation, grounding, and abstention designed in from the start rather than bolted on.
Measurement is the design tool
The best architecture decisions come from watching a system fail in specific, measurable ways. I treat evaluation as a first-class part of the build, so each layer exists because data justified it.
Capability plus adoption
An AI capability nobody uses is worth nothing. My work pairs building real systems with enabling the engineers and organizations around them — which is how adoption moved from ~20% to 95%.
Honest about limits
Trustworthy AI knows when not to answer. I'm precise about what a system has actually demonstrated versus what it might do at scale, because credibility compounds and hype doesn't.
A decade building systems and teams
Ten years across enterprise AI transformation, analytics, technical operations, software engineering, and instruction. Every role framed the same way: the problem, the scope, what I led, and the measurable impact.
Principal Technical Development Manager
Feb 2025 – Jul 2026ZoomInfoEnterprise AI transformationProblemAI capability was concentrated in experiments rather than production, with adoption stuck near 20% across the organization.
ScopeA portfolio of 68 enterprise AI initiatives spanning Engineering, Product, GTM, HR, Finance, Legal, and Operations across 10+ business organizations.
What I led- Designed, built, and deployed 15+ production LLM-powered AI agents generating 300+ hours of productivity gains per month.
- Architected an internal AI enablement platform with reusable workflows, prompt libraries, embedded assistants, evaluation frameworks, and deployment playbooks.
- Led Responsible AI certification for 368 engineers across Israel and India in partnership with Google Cloud.
- Built AI implementation frameworks spanning discovery, prototyping, evaluation, deployment, adoption, and measurement.
Impact- Increased AI adoption from ~20% to 95%.
- Enabled 500+ software engineers globally in LLM development, agent architectures, and Responsible AI.
- Connected AI capabilities to measurable outcomes in developer productivity and operational efficiency.
Director, Insights & Training
Mar 2022 – Jan 20252UAnalytics & organizational transformationProblemLarge-scale technical programs lacked the analytics and enablement systems needed to measure quality and act on operational risk.
ScopeEnterprise analytics, technical enablement, and workforce transformation across software engineering, cybersecurity, AI, and data science organizations.
What I led- Designed Python, SQL, NLP, and analytics solutions that cut operational audit time from 2.25 hours to ~10 minutes.
- Built analytics and performance-intelligence systems to identify operational risk and guide strategic decisions.
- Translated operational requirements into data workflows, reporting systems, and scalable processes.
Impact- Improved efficiency by more than 13x on operational audits.
- Raised educator quality metrics from 3.4 to 4.8.
- Reduced at-risk educators from 39% to 6% through analytics-driven coaching.
Senior Manager, Online Instruction
Nov 2019 – Mar 20222UTechnical operations leadershipProblemDistributed technical education programs needed scalable operations to maintain delivery quality as they grew.
ScopeGlobal software engineering operations supporting 140 instructors, 230 teaching assistants, and 90 mentors.
What I led- Built onboarding, technical enablement, coaching, and quality-assurance systems for engineering education teams.
- Used operational data and performance analytics to identify systemic issues and improve delivery quality.
- Partnered across engineering, operations, curriculum, and leadership to implement improvements.
Impact- Scaled distributed technical organizations while sustaining delivery quality.
- Established performance-management systems used across programs.
Full Stack Engineer / Mentor Manager
Feb 2019 – Nov 20192USoftware engineering & enablementProblemEngineering mentors and teams needed better onboarding, documentation, and developer workflows to scale their capacity.
ScopeSoftware engineering onboarding systems, technical documentation, and learning infrastructure for engineering teams and mentors.
What I led- Developed and supported full-stack applications using JavaScript, APIs, databases, and Git-based workflows.
- Built onboarding systems, technical documentation, and developer workflows for engineering teams.
Impact- Increased mentor capacity from 2 to 7 students.
- Maintained a 4.8/5 Net Promoter Score through improved systems and processes.
Instructor, Software Engineering & Data Science
Oct 2017 – Mar 2020UC Berkeley ExtensionTechnical instructionProblemEngineers and technical professionals needed hands-on guidance to build real applications and data solutions.
ScopeSoftware engineering and data science instruction covering Python, JavaScript, SQL, APIs, databases, analytics, and full-stack development.
What I led- Guided engineers through architecture, debugging, application development, and production-oriented projects.
- Mentored learners in translating technical requirements into functional applications and data solutions.
Impact- Prepared technical professionals for real-world engineering and data science work.
Let's talk about building AI that ships
Whether it's an applied AI system, an enterprise transformation, or the enablement that makes adoption stick — I'd welcome the conversation.