Yeah, I have never understood the over reliance on AI. Writing the code is not the challenge. The time it takes to push a new feature and test it out is often trivial, maybe a few hours.
The real challenge is forming the new ideas in the first place and most of those new ideas coming either from using the code as a product or time spent maintaining and refactoring large code.
Anyways, if you want to continue on the path towards regaining control and take it to the next level I wrote something similar here: <a href="https://blog.sharefile.systems/be-brave-go-low/" rel="nofollow">https://blog.sharefile.systems/be-brave-go-low/
Wriring the code is not the challenge, but it's what was taking up most of the time. Not the typing itself, but also because I had to think of how to implement it.
Now I can just say "add 2FA" and in 5 minutes, while I test something else, it is done.
It also made iterations a lot faster, you can try something out, see how it feels, if it doesn't work, you can just trash all the code and start again.
Haven't typed a line of code or read any code for over 6 months now.
And I used to love coding and be a competitive programmer, but this is how "coding" goes nowdays.
I have a mental model of what it would do, and how it would work, and I ask questions to confirm things and tell it to watch for specific gotchas. Then simply test the feature myself a bit.
Security-wise, I think the latest cyber models are better than me anyway at finding vulnerabilitates and pentesting such features.
Plus 2FA is a very common pattern, so it likely has in the training dataset many really good implementations.
You are abdicating your responsibility, which is fine for toy apps for yourself, but less fine when people expect that 2fa implementation to protect their accounts.
I think it's the opposite, with the right guideance, testing, frameworks, and using the top models today, implementing something with AI is most of the times better than what most developers would do.
I basically only manually test e2e myself, other tests are automated, code review is automated. I can tell the model to test for me too specific things or to add tests for specific potential issues, performance benchmarks, compared different implementations, etc.
The focus is a lot more around the code than on the code.
You don’t, that’s why in agentic world your codebase is only as good as what you can prove. For this reason, you’re going to see more languages evolving feature like capability permission, effect/coeffect types, refinement types, formal verifiers, strict type checkers, static analysis and so forth. Tests are only a small part of the verification. These ideas are old and have sat outside of the mainstream coding world, but the value proposition in the ai age has changed enough their relevance is renewed.
Not arguing for not reading test code. A lot can be achieved by instructing agents to balance out the testing pyramid with the right amount of fast end to end tests, property based tests for the right things, parametrized example based tests. Ensuring the local and CI has the right mix of tests running at right time. On projects where I have less time to review AI generated code I channel my anxiety into setting up guardrails and processes for the agents and then force them to bump up against them. In the end I view it as creating frameworks which allow me to outsource some of my attention to the agents, so that I can claw back some of that time to go set up more guardrails for more agents who are working on something else.
austin-cheney · · focus · HN ↗
The real challenge is forming the new ideas in the first place and most of those new ideas coming either from using the code as a product or time spent maintaining and refactoring large code.
Anyways, if you want to continue on the path towards regaining control and take it to the next level I wrote something similar here: <a href="https://blog.sharefile.systems/be-brave-go-low/" rel="nofollow">https://blog.sharefile.systems/be-brave-go-low/
XCSme · · focus · HN ↗
Now I can just say "add 2FA" and in 5 minutes, while I test something else, it is done.
It also made iterations a lot faster, you can try something out, see how it feels, if it doesn't work, you can just trash all the code and start again.
anygivnthursday · · focus · HN ↗
XCSme · · focus · HN ↗
Haven't typed a line of code or read any code for over 6 months now.
And I used to love coding and be a competitive programmer, but this is how "coding" goes nowdays.
I have a mental model of what it would do, and how it would work, and I ask questions to confirm things and tell it to watch for specific gotchas. Then simply test the feature myself a bit.
Security-wise, I think the latest cyber models are better than me anyway at finding vulnerabilitates and pentesting such features.
Plus 2FA is a very common pattern, so it likely has in the training dataset many really good implementations.
OccamsMirror · · focus · HN ↗
XCSme · · focus · HN ↗
I basically only manually test e2e myself, other tests are automated, code review is automated. I can tell the model to test for me too specific things or to add tests for specific potential issues, performance benchmarks, compared different implementations, etc.
The focus is a lot more around the code than on the code.
aprilthird2021 · · focus · HN ↗
But you didn't write or read any of the tests so how do you know they are accurate?
ModernMech · · focus · HN ↗
usewik · · focus · HN ↗