I enjoy creating content in different ways - writing, speaking at events, or being a guest on podcasts. If you have an interesting opportunity for me, please send me an email and let me know about it.
Our EKS cluster serves over 45,000 requests per second. For months, all good. Then the bill started climbing, with no deploys and no alerts. At this scale, how your services talk to each other really matters, and AWS charges for every word. Here's how we tracked it down, plus a few other networking surprises along the way.
CloudWatch has changed—and most people haven't noticed. In this talk, I'll dig into why the usual complaints about CloudWatch being clunky, overpriced, or too limited just don't hold up anymore. I'll walk through real-world ways to use its newer features to build solid observability setups—without jumping straight to expensive third-party tools. We'll also cover how to keep costs low without losing the metrics, logs, and traces you actually need.
Ever wondered why your monitoring setup isn't cutting it anymore? It's probably because you're not doing observability - and there's a big difference. I'll share my journey of transforming teams from "we monitor our systems" to "we understand our systems." We'll dive into real stories of how observability changed how teams work together, from developers writing code to ops folks running it. No buzzwords, just practical insights on making observability part of your team's DNA.