- Brief thoughts on the recent Cloudflare outage: The Cloudflare incident was interesting because there was a failure mode that had never been seen before and also a co-incident (the status page went down at the same time) that made the context harder to investigate. Another interesting note - protective systems design to help us can sometimes cause harm!
- Why Strong Consistency?: Eventual consistency makes it hard to reason about what a reasonable looking program might do given a common read-modify-write type update action.
- Why I Ignore The Spotlight as a Staff Engineer: I don’t think I’m a spotlight seeking person either. I’m happiest when I’m helping someone else do great things. And figuring out how to automate something. :)
- Why speed matters: A small idea with a lot of punch. Move faster so you can more quickly learn where the weak points in your approach are going to be.
- Cloudflare outage on December 5, 2025: Another big incident at cloudflare yesterday. That’s 2 in as many weeks. They have a global config service that replicates change very quickly without much of a promotion process it sounds like.
- Leaving Intel: Brendan Gregg has done a lot at intel. What a career and a pioneer of systems performance thinking!
