[flagged]
Using AI to write better code more slowly
391–400 of 511 posts
Re: Using AI to write better code more slowly
#392I've hit this point with AI where it's not a simple process, but a long drawn out back and forth. I'll use AI to design the implementation of a medium sized, cross cutting feature. Review all the details, maybe iterate on just that. Then implement with Claude 4.7 Max - which runs slower, but does a better job. Then review the implementation, then have Codex GPT 5.5 xhigh fast review it - which almost always finds cor…
Sounds exhausting
It is.. but so is dealing with issues at runtime, going through weeks of revisions, and dealing with technical debt.
Re: Using AI to write better code more slowly
#393Earlier quoted context omitted.
people said the same with any innovation
Did they? Genuine question, because I do wonder if people in some industries in the past were ever anxious about these specific things (especially skill attrition).
> I do wonder if people in some industries in the past were ever anxious about these specific things (especially skill attrition).
I've spoken with some people (now in their 60s & 70s) that worried about skill atrophy in their line of work.
First they worried about atrophy. Then they watched skill dry up. Now they know it's not available to buy anywhere. In the better cases the skills still exist, but entirely overseas.
These are people I could recognize as sharp engineers, even if I don't know their domains at all. I had to take them at their word about the value in what was lost. The problem is that it's easy to assume that business (or at least society) would prevent degradation of valuable knowledge over time.
Re: Using AI to write better code more slowly
#394This article doesn't address writing code with AI, just code review. My issue with agentic coding is that I make numerous micro-architectural decisions while programming. I almost never have a full spec up front and develop one as I consider what I am writing. When using Claude Code or Codex, that is all gone. Claude Code is extremely eager to reach the end goal to the point that it feels like a fever dream to write…
I find using the LLM to generate different git repo skeletons for the same class of project using the 4-5 different programming languages I’m familiar with is really interesting and helpful. Then I ask it to explicitly describe its design decisions for different parts of the small codebase, i.e. what do the internal APIs look like, so that if you make changes in one section of the codebase, you can be sure you don’t accidentally generate problems in another section of the codebase. Only once you’ve worked out all such constraints, clarified dependencies, etc. do you start generating code in each subsection and that’s done using the specific constraints for that section in each prompt, and reviewing all the code. This is also when you generate the tests for each subsection. Finally this is where using a different LLM(s) for code review after the code is written becomes important. It’s a slow process certainly but it seems to work pretty well.
Re: Using AI to write better code more slowly
#395I've hit this point with AI where it's not a simple process, but a long drawn out back and forth. I'll use AI to design the implementation of a medium sized, cross cutting feature. Review all the details, maybe iterate on just that. Then implement with Claude 4.7 Max - which runs slower, but does a better job. Then review the implementation, then have Codex GPT 5.5 xhigh fast review it - which almost always finds cor…
I think you need a skill to review those code by agent itself, but in a different role, not the one who wrote them. I did some research on this and developed a skill to get things done. By now it works well though I decide to prove and improve it with more tests. Dog food is not always delicious but not too bad either.
The problem is that I manually review the code before/after the review, as well as review the items to review themselves. You could easily put AI into a review infinite loop if you let it, and you also risk the code base going off the rails if you let AI go wild.
It's actually happened a few times where I need to back out entire features because AI went too far and I lost control/understanding of what the code is doing. Many people will give up at that point and let AI do everything - that is a mistake, at least right now and how you end up with unmaintainable vibe spaghetti slop.
Re: Using AI to write better code more slowly
#396As a junior, i do actually enjoy going back and forth with the AI discussing different ways to implement something and exploring alternatives. More often than not, I'd have an architectural idea that I'm not that confident in. The process of talking with the LLM takes a long time but it helps me sharpen the initial approach or even come up with a new one depending on the requirements.
Same. My total agentic session time/effort is about 70/30 discussion vs code-gen. I've got 12 years experience.
Re: Using AI to write better code more slowly
#397I've hit this point with AI where it's not a simple process, but a long drawn out back and forth. I'll use AI to design the implementation of a medium sized, cross cutting feature. Review all the details, maybe iterate on just that. Then implement with Claude 4.7 Max - which runs slower, but does a better job. Then review the implementation, then have Codex GPT 5.5 xhigh fast review it - which almost always finds cor…
I follow a similar approach and use multiple LLMs per task. The quality improvement is surprisingly large. Lately I’ve been experimenting with adding an explicit reward function so the models optimize for measurable output quality. This creates a generate, critique, revise loop where candidate answers compete for a higher score. It feels promising because it reduces the amount of handholding for every task. It is als…
I don't have it automated, but I score on minimizing lines of code added, readability of the code, and quality of the architecture.
Re: Using AI to write better code more slowly
#398As a junior, i do actually enjoy going back and forth with the AI discussing different ways to implement something and exploring alternatives. More often than not, I'd have an architectural idea that I'm not that confident in. The process of talking with the LLM takes a long time but it helps me sharpen the initial approach or even come up with a new one depending on the requirements.
This used to be called Pair Programming. And just letting you know, it’s not just juniors… we’re all do this :)
Re: Using AI to write better code more slowly
#399I've hit this point with AI where it's not a simple process, but a long drawn out back and forth. I'll use AI to design the implementation of a medium sized, cross cutting feature. Review all the details, maybe iterate on just that. Then implement with Claude 4.7 Max - which runs slower, but does a better job. Then review the implementation, then have Codex GPT 5.5 xhigh fast review it - which almost always finds cor…
tbh I'm just confused at why people ask AI to design features. Do you not know how to design a feature? Do you not know what you want? This stuff works so much better when you just tell it what to do
Oh course it's not black and white, there are many shades of grey in how detailed the design of a feature can be. Often even if I know low level details, I'll only give the AI high level requirements because I want to see how it would do it. Often it comes up alternative/better ways of doing what I planned and I incorporate those ideas into the final design.
Re: Using AI to write better code more slowly
#400As a junior, i do actually enjoy going back and forth with the AI discussing different ways to implement something and exploring alternatives. More often than not, I'd have an architectural idea that I'm not that confident in. The process of talking with the LLM takes a long time but it helps me sharpen the initial approach or even come up with a new one depending on the requirements.
Be sure to explicitly ask for critiques or alternatives. In my experience the machine is really susceptible to a sort of anchoring effect.