On giving talks I haven't given

Caches solve a problem and create three new ones: invalidation, observability, and the slow drift between what the cache thinks is true and what the source thinks is true.

I keep a small log of decisions in a text file. Not architectural — just small ones. "Picked X over Y because Z." Future me reads it more than I thought he would.

I used to add a layer of abstraction the first time I saw a pattern repeat. Now I wait for the third. The waiting is uncomfortable, but the abstractions that survive that wait tend to last.

There's a kind of premature optimization that comes from imagining a future team. "What if someone needs to swap this out?" Often, no one does.

I trust monorepos more in the second year than I did in the first. Most of the early friction goes away once the tooling catches up.