←back to thread

75 points throwaway-ai-qs | 3 comments | | HN request time: 0.021s | source

Between code reviews, and AI generated rubbish, I've had it. Whether it's people relying on AI to write pull request descriptions (that are crap by the way), or using it to generate tests.. I'm sick of it.

Over the year, I've been doing a tonne of consulting. The last three months I've watched at least 8 companies embrace AI generated pip for coding, testing, and code reviews. Honestly, the best suggestions I've seen are found by linters in CI, and spell checkers. Is this what we've come to?

My question for my fellow HNers.. is this what the future holds? Is this everywhere? I think I'm finally ready to get off the ride.

Show context
barrell ◴[] No.45279108[source]
I'm not convinced it's what the future holds for three main reasons:

1. I was a pretty early adopter of LLMs for coding. It got to the point where most of my code was written by an LLM. Eventually this tapered off week by week to the level it is now... which is literally 0. It's more effort to explain a problem to an LLM than it is to just think it through. I can't imagine I'm that special, just a year ahead of the curve.

2. The maintenance burden of code that has no real author is felt months/years after the code in written. Organizations then react a few months/years after that.

3. The quality is not getting better (see gpt 5) and the cost is not going down (see Claude Code, cursor, etc). Eventually the bills will come due and at the very least that will reduce the amount of code generated by an LLM.

I very easily could be wrong, but I think there is hope and if anyone tells me "it's the future" I just hear "it's the present". No one knows what the future holds.

I'm looking for another technical co-founder (in addition to me) to come work on fun hard problems in a hand written Elixir codebase (frontend is clojurescript because <3 functional programming), if anyone is looking for a non-LLM-coded product! https://phrasing.app

replies(3): >>45279156 #>>45279259 #>>45279390 #
james2doyle ◴[] No.45279259[source]
Totally agree. I use for chores (generate an initial README, document the changes from this diff, summarize this release, scaffold out a new $LANG/$FRAMEWORK project) that are well understood. I have also been using it to work in languages that I can/have written in the past but are out of practice with (python) but I’m still babysitting it.

I recently used it to write a Sublime Text plugin for me and forked a Chrome extension and added a bunch of features to. Both open source and pretty trivial projects.

However, I rarely use it to write code for me in client projects. I need to know and understand everything going out that we are getting paid for.

replies(1): >>45280211 #
bdangubic ◴[] No.45280211[source]
> I need to know and understand everything going out that we are getting paid for.

what is preventing you from this even if you are not the one typing it up? you can actually understand more when you remove the burden of typing, keep asking questions, iterate on the code, do code review, security review, performance review… if done “right” you can end up not only understanding better but learning a bunch of stuff you didn’t know aling the way

replies(1): >>45280878 #
1. barrell ◴[] No.45280878[source]
I have never met an engineer whose abilities were limited by their typing speed. Most engineers can already type far faster than they can critically think/store memories
replies(2): >>45281768 #>>45282398 #
2. bdangubic ◴[] No.45281768[source]
My abilities are limited by time, I don’t work (never have) more than 6-7 hours per day. Hence I automate everything that requires my time (and can be automated). If I have something that saves me 10 minutes per day, that is significant for me (should be for you too if you value your time). now imagine if I have something that saves me 90 minutes per day…
3. JustExAWS ◴[] No.45282398[source]
I don’t use Claude Code or any of the newer coding assistants. I use ChatGPT and tell it the code I want with the same type of specifications I would do when I am designing an implementation. It can definitely type 200+ lines of correct code faster than I could especially since I would need to look up the API calls for the AWS SDK.

I treat it like a junior developer.