This would sound insane to me from two years ago but I have recently started insisting on writing all my own commit messages and pull request descriptions. I do usually have an agent review them for factual accuracy, but not rephrase them.
It slows things down a bit, but in the best possible way. It has helped immensely to improve the depth of my understanding of the agent-generated code. When agents are doing everything its way too easy to “skim” diffs and not really absorb them.
I always prided myself on my technical writing, and commit messages and PRs were a great place to hone that skill. I found that I missed it and my work is better now I’ve reclaimed that part of my old job back.
Why do you want to understand the agent generated code though? That would make you the bottleneck. Shouldn’t you just concentrate on checking for optimal outcomes, exhaustively trying to find/prompt edge cases and profiling performance?
Letting agents run wild with a codebase and no humans understanding it is a recipe for disaster across so many axes.
You are not a serious person if you recommend that.
Check back in a couple years and I’m sure they’ll be there but today’s frontier models 100% absolutely are not ready to fully own a nontrivial production codebase with no human involvement.
I do check most of agent code, not because I really care about every tiny detail but to catch if it's going to shoot itself (and my project) in the foot. If I only check the result and it's OK, it still doesn't mean that in two features I won't totally blow up the app.
I'm a happy little bottleneck, what's wrong with that? How do you even know how to ask it to build stuff for you if you don't understand what foundations are you building on?
Writing commit message (aka. what has changed) gives me a sense of control that I know what's happening and I can confidently build the next thing.
semiquaver · · focus · HN ↗
It slows things down a bit, but in the best possible way. It has helped immensely to improve the depth of my understanding of the agent-generated code. When agents are doing everything its way too easy to “skim” diffs and not really absorb them.
I always prided myself on my technical writing, and commit messages and PRs were a great place to hone that skill. I found that I missed it and my work is better now I’ve reclaimed that part of my old job back.
dyauspitr · · focus · HN ↗
codechicago277 · · focus · HN ↗
semiquaver · · focus · HN ↗
You are not a serious person if you recommend that.
Check back in a couple years and I’m sure they’ll be there but today’s frontier models 100% absolutely are not ready to fully own a nontrivial production codebase with no human involvement.
intended · · focus · HN ↗
If your work has little liability then you can afford to not care beyond “does it work”.
If you have to worry about quality and ensuring you don’t get sued, you make sure the process works.
If it has to maintainable, you you need the mental model to be present in someone’s head.
tiborsaas · · focus · HN ↗
I'm a happy little bottleneck, what's wrong with that? How do you even know how to ask it to build stuff for you if you don't understand what foundations are you building on?
Writing commit message (aka. what has changed) gives me a sense of control that I know what's happening and I can confidently build the next thing.
FuckButtons · · focus · HN ↗