Speaking
I speak about behavioral contract testing for AI, AI-augmented engineering methodology, and the patterns I've developed across 20+ years of building test infrastructure from scratch. Interested in having me speak? Get in touch.
Zero Trust for AI: Behavioral Contracts for the AI Era
Keynote | 40-45 min | Live Demo
When your application uses an LLM, assertEquals is useless. The output is different every time. This keynote introduces behavioral contract testing — a pattern using custom Playwright matchers to validate AI output by testing behavior, not bytes. Maps the cybersecurity zero-trust model to AI output validation. Attendees see 8 custom matchers in action against live AI endpoints and leave with implementable patterns.
The Brain vs. The Body: AI Evaluation and AI Testing Are Different Problems
Breakout | 35 min
"AI evaluation" and "AI testing" are not the same thing. Model evaluation happens in Python, offline, against scored datasets. Product testing happens in a browser, in real-time, against live endpoints. This talk defines a new category distinction the industry hasn't made yet, and introduces the AI Trust Score — a combined brain + body metric.
680 Runs, Zero Retries: Statistical Confidence in Test Engineering
Breakout | 30 min
"It passed once" is not a fix. This talk covers a systematic approach to flaky test elimination: root cause pattern classification, 680-run stress validation, and predictive hardening — fixing failure classes before they fail. Resonates with any team that has CI reliability pain.
CI/CD Is Lying to You: 8 Pipeline Failures Nobody Noticed
Breakout | 35 min
An anthology of real CI/CD failures — from 13-day silent failure streaks to 95% speedups with 15 lines of bash to GitHub Actions that phone home to commercial APIs. Each story follows discovery, root cause, fix, and metric. Your pipeline is giving you false confidence.
Your AI Wrote the Code, But Who Audited the Dependencies?
Breakout | 30 min
AI coding assistants introduce dependencies based on training data popularity, not security posture. This talk shares a framework for auditing AI-introduced dependencies across direct risk, transitive risk, geopolitical supply chain, and single-maintainer exposure.
Upcoming Events
No confirmed events yet — CFP submissions in progress for 2026/2027 conferences. Check back soon or reach out to book me.
Book Me for Your Event
Available for conferences, meetups, workshops, and podcasts. All talks include live demos and actionable takeaways.