Hacker News
points by monsieurbanana 3 days ago | 0 comments
I guess I have to temper my own statement a bit. I do see situations where I would generate and send automated LLM messages. Probably to reply to generated unsolicited messages or pull-requests.

But to upload to debian new's queue you already need to have pre-approval, it's a different situation from e.g. a public git repository that anyone can push to.

[–] g-b-r 3 days ago
> where I would generate and send automated LLM messages. Probably to reply to generated unsolicited messages or pull-requests.

Why would you use an LLM instead of a fixed message?

[–] monsieurbanana 3 days ago
To help classify between human and llm-generated ones? I don't know, and you're probably right than in this case a fixed message would be better.

Another situation where I would use llms to generate messages is if we've mutually agreed to do that. For example automated review messages at work, if we've agreed that it would be a part of our workflow.

A good rule of thumb might be unsolicited vs solicited llm messages.

[–] g-b-r 3 days ago
Ok, for unattended replies indeed a fixed message is much better; you can't predict what an LLM will output, and a certain impersonal message is better than a personalized one that might be completely senseless.

If you mutually agreed of course it's fine

[–] tetha 3 days ago
And even with a mutual agreement - default LLM output tends to be very hard and inefficient to read. We've spent quite a bit of time at work to setup a few writing skills for Claude.

For example, mine is a technical writing skill based on the old book "The Art of Scientific Writing": Front load information as much as possible - the first paragraph of a document should be a summary, the first sentence of a paragraph should summarize. Short sentences building on each other. Always show evidence.

Someone else spent quite a while optimizing a standardized PR description template, with a section for intent, expected changes, risk, rollout coordination and such.

That made the slop a lot more tolerable.

[–] InsidiousWombat 3 days ago
But there is also way to throttle that. Have AI go thru submission guidelines, compare it to PR and have it shut up when most (not all! AI will always give a reason if you ask it to check something) of it is correct, and only answer when the PR itself doesn't comply to the project rules