▲ 48 points
back
21 comments
This misses the best use of AI in my opinion, which is to gain understanding. Whst is this bit of code doing? Is there a risk of data leaking here? Are permissions enforced downstream of this function?
Code review can go much deeper now if you use AI to aggressively attack a PR combined with your human insight. Same for planning a feature:
Can I consolidate this logic to a shared function? Does the error surface to the user and are there any gaps? What preexisting functionality is affected by this PR? Can this query be made more efficient?
LLMs are great with focused questions, up and down abstraction layers and across all kinds of concerns. Stack up these focused concerns into a rich understanding of what you are doing or writing.
Understanding is the real output, code is the byproduct.
> LLMs are great with focused questions, up and down abstraction layers and across all kinds of concerns. Stack up these focused concerns into a rich understanding of what you are doing or writing.
I agree. This is much better advice than what the article tries to do, which is to give best practices around how to use or not use AI to code.
> This misses the best use of AI in my opinion, which is to gain understanding. Whst is this bit of code doing? Is there a risk of data leaking here? Are permissions enforced downstream of this function?
Do I understand when I just let AI regurgitate the function and behavior of the code?
Wish I could upvote twice or something
Exactly, i so wish the narratives around LLMs in most fields were around how they can enable us to be better ourselves...
SOMEONE didn't read the article
> You code and you have the AI check your work. Treat it like a reviewer. That’s my advice in a sentence.
Bro you're giving the same advice as the article
>Preferably, don’t give the AI access to the codebase. Copy-paste snippets of your code in the web interface.
Well, I can't ask it for an overview of a system it can't see.
> Preferably, don’t give the AI access to the codebase. [...] If the AI does have access to the codebase, this access is read-only.
also
> I don’t follow all these religiously.
So if you need an overview of a system, you can and should provide access to the entire codebase. Even the rule itself says "preferably", not "don't ever". If you have different needs, you can always just look at it as a useful guideline. It's just a blog post by some rando on the internet, not the epiphany of AI coding's Ten Commandments.
This article hits very close to what I have come up with myself and its good to see others thinking the same lines.
Core modules: Coded by myself, AI reviews and AI to discover/learn.
Stuff I don't care about Craft: API layer, CLI layer, Smoke tests, Integ tests - Dial AI heavy, and lighter human reviews accordingly
“But that would be like telling a student in the 70s to pretend that calculators or computers don’t exist.”
If students of the 70s or today pretended they didn’t exist up to a certain point when they needed them to move forward, like bioinformatics or something, they 100% would be better off. There is plenty of research on off loading thinning providing a worse understanding of the material - eg side rules proving a better understanding than calculators.
“You can use the electronic tuner once you learn to tune by ear” - any decent guitar teacher
Exactly. I always think about the ways consumers make terrible decisions with everyday purchases at the grocery store etc. If people could easily estimate unit costs in their head for example, a lot of predatory options would have to go away. But calculators keep people from having to get good with everyday math in their head, while those same calculators sit unused in many everyday scenarios where people don't go to the effort to use them.
Quick estimating (which is a different skill from calculating exact answers using pen and paper) is most useful there. People do carry computers with them these days and they do have calculator apps. Some shops I buy from do have unit prices for many items.
I think AI might be more of a problem, at least for some skills because it is not just doing a small unit of work. its more like the problem of doing more complex calculations. An example I have come across with calculators is students learning statistics without knowing how to calculate a variance, because they just put the numbers into a calculator. There is a failure there to learn the concept - they do not really know what a variance or SD is. I can imagine AI doing a lot of that.
I don't know if this was universal, but when I was a student we weren't allowed calculators until I believe 10th grade
You learn stuff by heart, so that when you mess up with the calculator you have a feel that stuff has gone wrong
Insightful:
> the term “vibe-coding” suggests a kind of laissez-faire attitude where you don’t really care about the outcome and you’re just having fun. That’s what the phrase meant when it was coined, but the world has moved on. In many companies professional programmers are using AI in such a way that it’s impossible to imagine that they are also reading the resulting code in detail. This is what modern vibe-coding is. Deferring to the AI, not worrying about the individual lines of code, and keeping an eye on whether the code passes its tests and throws up any problems in production.
This way of working is the only one that justifies the trillion-dollar bet on the AI industry. I agree, this method should be called vibe-coding.
It should depend on whether you've taken other steps to constrain the output of the LLM. I've been building some pretty elaborate test harnesses (which I test independently beforehand), to ensure that I get something that does what I ask of it.
I don't care which words we use, but they should separate this sort of empirically-minded work (versus the rationalism of typical development) from just dropping a few sentences into the prompt and crossing your fingers.
> empirically-minded work
Yeah, I understand that this testing-oriented constraint method is what AI coding enthusiasts are pitching. Sure, it's different from pure vibe-coding as originally defined, but nobody is seriously suggesting professionals should do pure vibe-coding anyway. It's a strawman. The word fits.
>It's a verbose coder, it overcomplicates things, and it works fast. Even if you had an inhuman level of attention, you couldn't keep up.
I wonder if that trick about prompting it to write like a 5th grader* would help here. Keep it simple!
*A trick which allows it to pass for human 70% of the time...
I went from just coding manually, to using LLMs for "surgical edits", to "woah, AGI!", to "haha whoops, not even close", to just coding manually (my brain still works!), to "surgical edits" again.
My current approach is "ask for very small diffs" + "review them very carefully".
I'm not working a job though, I'm working on a multiplayer game.
Main findings: The frontier models can't reliably modify Pong without breaking it, so their skill appears to be quite domain-specific. (OK, to be fair, neither can I half the time!) This is probably because they are "time blind". I had one model try to test a game by running it at 0.1 frames per second and shoving each frame in the vision API...
If you leave any room for a misunderstanding, they will laser in on do it and do the stupidest thing possible. If you're not checking everything carefully, you will discover this later, and you will cry.
Formal proofs, oddly enough, do not improve the situation: they will simply prove mathematically that the absurd and pointless and backwards implementation is completely without defects. (It obviously does help within an implementation, though.)
They can't formally prove what the hell you meant when you told them to build something. That job remains frustratingly human!
Current dissatisfaction: (1) Harnesses are designed for super bloated codebases (i.e. designed to load as little context as possible) which make them pretty clunky for small repos and small edits. (I had a Surgical Edit Tool I need to bring back...), (2) Current LLMs are anal about verifying the most trivial change, even without prompting, even if it's impossible for them to verify it because they're blind so they start measuring pixel data in Python... Both of which eat up Speed and Cost, taking the work even further from Realtime/Interactive to Tedious/Sad.
I think one of the best use cases of AI is as a natural language interface to programming. The syntax of a programming language is an opinionated part of its design that often adds to the complexity of learning the language.
Very interesting. I find this is specially applicable to people learning or perfecting coding skills, not so much for developers already proficient in coding. What do you think?