“Principal Architect | 18+ years in performance engineering, cloud scalability and observability for mission-critical platforms.”
I've spent 18+ years answering one question: what actually happens to software when millions of people use it at once? Most teams find out the hard way — in production, on the worst possible day.
As a Principal Architect, I lead performance engineering and observability for cloud-native platforms serving millions of users. That means building automated performance gates into CI/CD so developers catch slowness in minutes instead of after release, tuning caching and database layers to take response times from seconds to microseconds, and deliberately breaking systems through chaos engineering so they heal themselves when something real goes wrong. I've engineered platforms through some of the highest-pressure traffic moments in consumer tech, including a Super Bowl product launch.
I'm a good guest because I speak in specifics and I'll disagree with the popular answer. Scaling horizontally is often the wrong move. Most observability projects end in dashboards nobody opens. And the majority of performance problems aren't code problems at all — they're design and culture problems wearing a code costume.
Recently I've been working where performance engineering meets GPU-heavy AI workloads, which is quietly invalidating a lot of assumptions the industry still relies on.
I'm AWS Solutions Architect certified, a certified chaos engineering practitioner, and completed a postgraduate program in AI and machine learning at the University of Texas at Austin. Your listeners will leave with things they can test on Monday.
