Joined 7 years ago
Those things are good enough to solve novel math problems I've never even heard about, surely it can figure out how you like your commit messages and apply it a million times over in the time it'd take you to make yourself a cup of coffee.
> What do you mean with "manually"?
I type `git rebase -i head~15`, I select the commits I want to include, rearrange lines manually, mark the ones that I want to edit, squash, or reword. I save the buffer, exit, git starts giving me individual commits to edit/reword, so I edit chunks for manually type in the reworded messages etc
These days I just tell LLM what I want and it does git for me as a byproduct and it looks more or less the way I want, except with like 1% typing on my part and 100x times faster.
It'd be weird to ask for evidence of that statement cause it's very generic and there's like a million instances of newer cars that are shit and old cars that are brilliant. But in general the statement totally holds and you'd need to be insane to be arguing otherwise.
Having 27 years of driving cars experience would add weight to the opinion stated, but otherwise not provide any substantive credible evidence.
I only have 17 years of experience and I fully agree with the author here. This is my personal opinion and I would not be fucked to respond to any demands for evidence since those demands are not meant to be satisfiable in the first place.
Like:
— My new car is much better than my old car — Hurr durr, how can you make those confident statements without showing any evidence, huh?
It's not that deep.
Like, at my job we went from wild-west everyone uses their own commit message style to very nice uniformed one now when everyone's using Claude Code following the style enforced by AGENTS.md/CLAUDE.md
Now I just type into the LLM what I want it do with git and it does it for me, much much quicker. I did not find "babysitting" git painful before, but I today I do feel like doing it manually is a just huge waste of time. A $20 LLM subscription could do this shit for me just as well or better while I spent my time doing more fun and interesting things.
Anthropic ships models with a helpful one-liner tag that makes the model hierarchy obvious. I think it wouldn't hurt if OpenAI did the same.
Having read both articles I'm in the opposite corner. I found Andrew's personal ad-hominem filled emotional post distasteful and Jarred's purely technical and filled with praise of Zig well written and in good taste.
I hate the messenger despite loosely agreeing with the message, I guess.
In practical terms it often means that you can give the agent a ticket and it'll work on it for several hours until the PR is ready. Occasionally it'll run into some question that it needs you to clarify something or make some directional choice, but overall it's pretty autonomous.
Because this takes so long, you normally run 3 or 4 of those in parallel.
Then between shepherding those session, you also run several PR review sessions at the same time. Those also run for quite some time when doing deep investigations.
And then you also have long running discovery/design sessions for various other projects or problems you're working in between.
I've tried Cmux, but it didn't do it for me, since the agent statuses were displayed on the workspaces and having multiple agents in the same workspace would sometime produce confusing results.
I've been using Herdr since the start of the week and so far it's been the best in terms of visibility of what my agents are doing and which of them need attention. The only wart I've noticed so far is that the performance is not always great — sometimes I see the text appearing with a noticeable delay as I type it.
