My hot take from chatting to @poteto is that we should use MORE abstractions in the AI age
You can use them (combined with harsh lint rules) to reduce the design space available to the agent and constrain them only to good decisions.
Combined with the fact that high-leverage abstractions let you do more with less code - so, more token efficient.
Plus, unwinding the damage from a bad abstraction is much cheaper with agents.
This runs counter to a lot of folks thinking that agents just want to read the raw code. They can, but they're not maximally efficient that way.
Be braver! Design abstractions.
@IAmManware DHH was so against AI at one point but his views have changed completely. He has a lot better judgement than us. Believe it or not, there will be far less jobs than we have now.
"we'll use a NoSQL database so we don't have to worry about schema migrations."
- someone who is about to spend the next two years reinventing relational joins in application-level javascript.
We're adding support for AGENTS.md to Claude Code.
Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md.
You can toggle this behavior in /config.
@arindam___paul blackpill take:
anything produced by agents has to be consumed through agents or you are ngmi in the slop tsunami
it's also why there is no place for avg talent anymore - armed with agents they now slow everyone else down instead of speeding them up
@IAmManware Whenever I start brainstorming the architecture for my side project, claude does most of the things and then a time comes where I think it ain't worth it if claude can do it all. Am I doing something wrong? How do you write code now? What should be the approach according to you?