I was blown away how clean the code for LSM compaction in TigerBeetle is, despite the complexity of callback model. Then Joran Dirk Greef told me that they rewrote it once and are will maybe rewrite it again because they're totally happy with it. That's commitment to code quality!

Ok, but there's some banger papers out there. Ngl

I turn 30 next week and in lieu of other plans I'm doing a little fundraiser! If you've ever enjoyed or benefited from my writing or communities, consider a little present. :) I've kicked things off by donating $300 each to Scholarship America and Malala Fund.

I agree. Recently experiencing this a lot, having to follow through layers of "helper" functions and abstractions only to realize it could've avoided intermediary allocations or just be a few LOC once inlined.

started using "chat" when talking to myself internally

No longer 3 bytes old.

Probably the smoothest conference I've attended so far. Was also great meeting old and new faces alike. Props to TigerBeetle again for setting it all up. #systemsdistributed

Blog: You can build any blocking sync primitive using any other one. And it's not just a neat realization. It actually works kprotty.me/2025/07/31/syn…

Writing fast code under optimizing compilers feels similar to *prompting: 1) Developing an intuitive understanding of how the model combines your patterns to outputs 2) Knowing what a "good" output is, in regards to correction. * any deterministic ML inference with text input

Dominik Tornow x.com/kingprotty/sta…

Blog: Batched Critical Sections A better alternative to mutexes kprotty.me/2025/09/08/bat…