Writing code was never the bottleneck in software engineering. LLMs made it cheaper to write, not easier to understand, review, or maintain. That’s still the bottleneck. Let’s not pretend it isn’t.
@sudo_goreng What about companies with real customers that, instead of improving the things customers complain about, are more interested in playing the AI game?
Customers want useful features, great UX, superb reliability, and fair pricing; they don't give a fuck if you're tokenmaxxing, they don't even care whether it's a human or a robot writing your code.
@sudo_goreng You need to see a lot of awful stuff, agree that it's simply awful, and do better. People who don't develop that still have bad taste, even with a lot of experience.
Hard disagree. Good taste is a lot more than just experience. There are a ton of experienced engineers with bad taste, and adding more experience won't simply change that.
Database internals are such a cool thing to talk about. I gave a talk about Merkle trees in 2019, and I thought I was talking about something so old that no one cared about it. Fast-forward to 2016, and they're still rocking!
Merkle trees are everywhere:
- ZFS uses them to detect data corruption
- Git uses them to verify repo integrity
- Cursor uses them for codebase sync
- Bitcoin uses them for transaction verification
Talked through how they work on the latest database stream.