Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy The thing that stumps my coworkers when they paste responses from LLMs to me is me asking for evidence it's correct.
Me: "Where's the corroborating evidence? Can you link me to the documentation that supports this?"
Them: "Huh?"
Me:
-
@abstractcode @daisy our team uses AI coding assistants a fair bit but if anyone ever said "the AI did it" in a PR review those would be grounds for an instant rejection of the PR.
There's an ongoing debate on whether these actually are useful or worsen the codebase in the long term but code ownership has to be non-negotiable for any of this stuff to work.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy this also seems extra toxic for open source: nobody releases a project so random strangers can try to score internet points by prompting a chatbot “for you”.
-
@abstractcode Yeah it's highly dependent on the codebase, the complexity of the problem and how "common" the stuff you're doing is. Had zero success in one org with niche tech but have found it useful and productive in another org with stricter base architecture and a good set of agent rules.
The unpredictability of when it's useful is the big problem though: building a mental model based on non-deterministic outcomes is basically impossible.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy this also pretty much summarizes the issues with LLM in education. When grading assignments the feedback no longer helps the student to understand a topic, it just teaching them how to write prompts to satisfy their professor‘s grading scheme.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy BIG SAME. It's worse when they respond back with some LLM message challenging you. - Or my favorite, we had a contractor tell me point blank they didn't use AI...
I've been trying to figure out a good way to handle this at our company. But so far it seems whatever guidelines I put in place are either fed right into their prompt or just ignored. No one seems to understand the code they're pushing off to a reviewer either.
-
@abstractcode @daisy this is so real. They use it to code for them, and take the blame for them too
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy you’re expected to use your time to review when they didn’t use their time to code the thing, and the “solution” is to make the LLM review itself, Ouroboros style
They want to vibe the requirements of product, vibe the design, vibe the implementation, vibe the tests, vibe the documentation, and vibe the reviews. They’ll reach a point in which they’ll need to vibe the customers, and maybe then they’ll realize how idiotic the whole thing is.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy it's similar with translations. It used to be recognising the pattern/misunderstandings which made mistakes easier to find, you would correct it and give feedback.
AI/LLM doesn't understand the text it's translating, mistakes are random, no feedback, no learning on the other end + it takes more time to correct everything if it's even possible to correct it (and it's paid worse)
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy my main issue is I keep going "did you actually look at this? "
The issues aren't things being slightly wrong in ways that the other developer might not be aware of, shit is just weird
Like a function that doesn't do any async operations returning a Task just like for fun I guess
-
@daisy this also pretty much summarizes the issues with LLM in education. When grading assignments the feedback no longer helps the student to understand a topic, it just teaching them how to write prompts to satisfy their professor‘s grading scheme.
@hannorein hang on. What? How is that a problem? If they were going to cheat, they would have found a method to do so anyway. So to say, students setting themselves up for failure is entirely not a new thing; students have always found methods to circumvent their systems. For example, how many students would ever be trusted by their professors to practice their learned trade? Not many, if any, according to anecdotal evidence or unofficial surveys of STEM professors.
-
@daisy you’re expected to use your time to review when they didn’t use their time to code the thing, and the “solution” is to make the LLM review itself, Ouroboros style
They want to vibe the requirements of product, vibe the design, vibe the implementation, vibe the tests, vibe the documentation, and vibe the reviews. They’ll reach a point in which they’ll need to vibe the customers, and maybe then they’ll realize how idiotic the whole thing is.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy A few months back I was doing a code review and mid file found a block of text stating something to the effect of “place rest of code here”. Just middle of the function, like it was meant to be code. And of course it wouldn’t build. And this is even when we have a corporate policy of not sending our code to an AI… It’s so exhausting.
-
@abstractcode @mattdarveniza none of it matters you dont use the fascism machine BECAUSE its a political project to devalue the one thing your paid for not because of how good or bad it is at your job
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy If you don't like it don't waste your time on them, ban AI slop from the project
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
There seems little doubt that the proliferation of this slop will result in enormous problems down the road that will be difficult and costly to clean up.
When they do, it SHOULD teach company executives to scale back if not stop the use of LLMs in their IT development shops, but, based on my experience of the lack of mgmt interest in quality improving systems analysis & design tools & methodologies back in the day, I fear it won't. -
There seems little doubt that the proliferation of this slop will result in enormous problems down the road that will be difficult and costly to clean up.
When they do, it SHOULD teach company executives to scale back if not stop the use of LLMs in their IT development shops, but, based on my experience of the lack of mgmt interest in quality improving systems analysis & design tools & methodologies back in the day, I fear it won't.@joeinwynnewood @daisy The management line will be "Well, you've not been sufficiently subservient to the AI so of course it's not helping you as much as it could. We'll hire someone who understands how to Tamagotchi these things appropriately. Good luck out there!" <door-slam/>
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy — oh, hell no.
I was thinking in terms of the rage I felt when someone committed and stomped over a bug I spent *hours* tracking down & testing and fixing. I think if I fix it, commit, and some viber + AI commits an entirely-new codebase (as prompt-written source is prone to be) needing that repeated, they’ll be dead in the desert by dawn.
-
@hannorein hang on. What? How is that a problem? If they were going to cheat, they would have found a method to do so anyway. So to say, students setting themselves up for failure is entirely not a new thing; students have always found methods to circumvent their systems. For example, how many students would ever be trusted by their professors to practice their learned trade? Not many, if any, according to anecdotal evidence or unofficial surveys of STEM professors.
I agree. I mean, before LLMS we were also just trying to deliver whatever the prof asks of us.
-
Genuinely one of the worst things for me about the vibe coding apocalypse is that it is steadily eroding my patience in code review.
It used to be that if you identified issues with someone’s code, you could explain why, and help your coworker learn and grow as a professional. And sometimes they’d respond by explaining why they did it that way, and then you get to learn and grow as well.
Now a lot of the time when I do code review, I feel like I’m not actually investing my time in learning, just giving them something to copy paste into an AI chatbot without engaging with either the code or the feedback.
@daisy but the velocity is bewildering. it used to be that an entire engineer team made an unmanageable spaghetti over a year from the code base. now it can happen in a single week by a single engineer!
-
E em0nm4stodon@infosec.exchange shared this topic
