undefined | Better HN

0 pointsLordGrey10d ago0 comments

> ... I make performance improvements, I stabilize CI, I improve code readability, remove compiler warnings, you name it ...

These are exactly the kind of tasks that I ask an AI tool to perform.

Claude, Codex, et al are terrible at innovation. What they are good at is regurgitating patterns they've seen before, which often mean refactoring something into a more stable/common format. You can paste compiler warnings and errors into an agentic tool's input box and have it fix them for you, with a good chance for success.

I feel for your position within your org, but these tools are definitely shaking things up. Some tasks will be given over entirely to agentic tools.

0 comments

tenacious_tuna9d ago

> These are exactly the kind of tasks that I ask an AI tool to perform.

Very reasonable nowadays, but those were things I was doing back in 2018 as a junior engineer.

> Some tasks will be given over entirely to agentic tools.

Absolutely, and I've found tremendous value in using agents to clean up old techdebt with oneline prompts. They run off, make the changes, modify tests, then put up a PR. It's brilliant and has fully reshaped my approach... but in a lot of ways expectations on my efficiency are much worse now because leadership thinks I can rewrite our techstack to another language over a weekend. It almost doesn't matter that I can pass all this tidying off onto an LLM because I'm expected to have 3x the output that I did a year ago.

j / k navigate · click thread line to collapse