←back to thread

Andrej Karpathy: Software in the era of AI [video]

(www.youtube.com)
1480 points sandslash | 4 comments | 19 Jun 25 00:33 UTC | HN request time: 0.533s | source
Show context
abdullin ◴[19 Jun 25 07:03 UTC] No.44316210[source]▶
Tight feedback loops are the key in working productively with software. I see that in codebases up to 700k lines of code (legacy 30yo 4GL ERP systems).

The best part is that AI-driven systems are fine with running even more tight loops than what a sane human would tolerate.

Eg. running full linting, testing and E2E/simulation suite after any minor change. Or generating 4 versions of PR for the same task so that the human could just pick the best one.

replies(7): >>44316306 #>>44316946 #>>44317531 #>>44317792 #>>44318080 #>>44318246 #>>44318794 #
1. yahoozoo ◴[19 Jun 25 11:15 UTC] No.44317531[source]▶
The problem is that every time you run your full automation with linting and tests, you’re filling up the context window more and more. I don’t know how people using Claude do it with its <300k context window. I get the “your message will exceed the length of this chat” message so many times.
replies(3): >>44317624 #>>44317703 #>>44318129 #
2. diggan ◴[19 Jun 25 11:31 UTC] No.44317624[source]▶
I don't know exactly how Claude works, but the way I work around this with my own stuff is prompting it to not display full outputs ever, and instead temporary redirect the output somewhere then grep from the log-file what it's looking for. So a test run outputting 10K lines of test output and one failure is easily found without polluting the context with 10K lines.
3. the_mitsuhiko ◴[19 Jun 25 11:43 UTC] No.44317703[source]▶
I started to use sub agents for that. That does not pollute the context as much
4. abdullin ◴[19 Jun 25 12:42 UTC] No.44318129[source]▶
Claude's approach is currently a bit dated.

Cursor.sh agents or especially OpenAI Codex illustrate that a tool doesn't need to keep on stuffing context window with irrelevant information in order to make progress on a task.

And if really needed, engineers report that Gemini Pro 2.5 keeps on working fine within 200k-500k token context. Above that - it is better to reset the context.