I've never been more informed or understood more about my code, pipelines, and stack than now- and its 100% due to AI reasoning about my projects, and me making an effort to learn.
For example, when studying maths or physics, it is very common to feel like you've understood everything, until you get to the exercises, and need to apply that understanding, and it's only then that you consolidate the knowledge and begin to truly understand in depth.
I like AI enhanced coding, but I do sometimes worry that we're not getting enough of that depth anymore.
Would you mind clarifying if this includes code, pipelines, stacks, etc... on which many other people are simultaneously working on in a group setting? Or are these all things that you alone are working on as an individual?
My experience has been that AI makes me tremendously more productive as an individual working alone. Put 20 people (all empowered by AI) on a project however, and it quickly falls to shit.
Now, to actually get the AI to do anything complex, it's basically required to both write docs and write tests because solidifying behavior only works when there's AI tests it can run to verify some behavior I've verified once.
Then there's things I'm never going to remember the AI had to do; recenly it was "blitting" to /dev/fb0 and it was trying to remember which encoding was which and what order, etc. Things I simply do not want to have to get a detail account of nor is it something I need to remember above the statement "some video drivers have their own RGB, BGR encoding standards" and "an image has a bit depth and size" etc.
These things I would have learned doing it myself, but I would also had to learn where it all breaks and how to interoperate between pillow and video drivers, etc. If I ever do this again, i'll still point the LLM at it with or without this library, and it'll do the same abstract reasoning.
So my knowledge is compact because I don't need to know the implementation for this video display code; I just need the tests and docs and I'll point the LLM at it if I need to extend it.
At least in my circle it's the classic leads/seniors (that predate the scrum "everyone's the same" thing) that are managing to get the most real mileage out of this.
No. There are multiple studies showing that skills atrophy is an actual thing. If your brain does not turn off while you are vibe-coding, keep it up and it soon will.
You have too, because that's the definition of vibe coding. If you use an LLM to assist you, but still keep your brain on, that's not vibe coding.
That's _your_ definition of vibe coding.
If that's not "turning your brain off", what is?
The property Scotsman is existing a priori and then a causality to behaviour is assumed. But here the term is defined by behaviour.
Of course, you absolutely can build a No True Scotsman argument on top of that distinction, but I don't think that's what GP was doing.
Is it possible to use it in a more involved way? Certainly. I try to do that. But it's challenging, because ultimately even taking a more collaborative approach, this thing puts out a lot of slop, and then I have to deal with going over the output and just spending most of my day in code review mode vs an author that doesn't ever learn from feedback.
The most frustrating thing for me beyond the feedback black hole is that when things do inevitably go wrong, trying to point out the concrete issue and instructing the LLM to do better around it is challenging; a lot of the time directives (even those in a global CLAUDE.md!) are just ignored either outright or as the context grows, or fail to be passed down to subagents; negative prompts are discouraged because of the pink elephant effect, so I have to jump through hoops to try and frame a "never do $thing" constraint as an affirmative prompt (which is then often ignored). That aside, english is a godawful language for specifying things compared to programming languages. It's just a really frustrating experience overall.
"Vibe coding" to me and others is just going "claude program me a wife that didn't leave with the kids and make no mistakes" and just rawdogging the output.
but that's just me.
Actually, for more complex work I think it's pretty common to spend a long time crafting some elaborate prompt, and then arguing with the agent for 10-20 turns, getting a "passable" plan, accepting it then arguing with the agent every step of the way because it's doing it wrong. It's incredibly frustrating, even with models like Fable.
Even after that, I'll often find out during later sessions that some feature that was supposed to be deprecated was actually silently left in "because I wasn't sure you wanted it completely gone" or something.
I agree that exploring a codebase through prompts is actually quite nice! But the mental model you get from that is almost always warped - you have to supplement that with reading the code, where 9 times out of 10 you will find some kind of discrepancy that you're unhappy with.
Why not optimize that process?
Splitting problems into smaller problems is huge. You want a concrete idea that you still own. Only tell an agent to do something you know it can nail - this will likely be one subsystem. The subsystems and how they talk is on you.
I'd really like something that works more like pair programming. Where you share an editor session with the AI, and it's much more interactive. eg. It says "I'm thinking about doing X here, what do you think?". Then you give a response and it continues. Or you can see it doing something silly and immediately stop it and correct something either with a prompt or by manually editing yourself, before letting it rip again. Or just ask it "why are you doing it that way?". Maybe with some control over the speed it's going, for when you feel confident about what it's doing.
Claude has made some positive changes over time where it asks you more when there are different approaches it can take. But I feel like I'm not being brought along with my mental model as much as I would like. A lot of the time it spits out a whole pile of code and then I have to go and build up the mental model after the fact, and correct a lot of what it's done.
The current approach is good a lot of the time though, when you're not really changing anything architectural and just want it to bash out code while you do somehthing else.
That sentence is probably the key.
There is zero chance of this stopping overall.
Much like with artists 90% of the commercial stuff - copy/ads with generic guy in suit picture - will be AI that is "good enough".
Less concerned about the code quality and more the labour market dynamics. If this goes anything like translation & creative space did then there is going to be a brutal shakeup where competition for remaining "real" seats gets intense
I’ve been gradually moving test coverage up, updating packages with CVEs (even if in some cases it’s more like a migration), fixing more bugs, writing load testing solutions, fixing badly made architectural choices and recently even migrated a codebase away from Oracle onto PostgreSQL and figured out some pretty good settings for the DB itself with the aforementioned test tool. Oh and introduced Testcontainers to actually do proper DB tests.
None of this would have happened without agents that churn even while I sleep, but somehow that is still more work cause I have to iterate on their output a lot and come up with additional ProjectLint rules whenever something new comes up. It’s like burnout².
Eh, wouldn't be sure. Every time I see slop in such a context, it makes my day worse. And usually ensures I wouldn't go to that business, or at least unconsciously biases me against it.
I scanned the article, never made it to the bottom, didn’t find the point the author was trying to make.
Howabout just learning to code better as a human without using AI. Just like learning a craft. Just refuse to use AI. Not possible because of peer pressure? Well then stop worrying and continue to vibe.
Either stop or stop complaining but don’t make excuses or write long articles that ramble and rant … sorry but I don’t understand what’s so hard about stopping.
I'll add this link to the pile: https://martinfowler.com/articles/exploring-gen-ai/sdd-3-too...
> spec-kit created a LOT of markdown files for me to review. They were repetitive, both with each other, and with the code that already existed. Some contained code already. Overall they were just very verbose and tedious to review. [...] To be honest, I’d rather review code than all these markdown files.
The hardest part of any of this extraction is that modern code is already an extremely dense representation of how the computer should work. You mostly can't change the code without changing behavior.
I bet Scryer works for his use case, and it's a joy to dogfood. I also bet it fully breaks down the moment a 2nd developer, who cares about different things, joins the team.
Intent is more natural to us than code, for the same reason assembly is more natural than binary.
1. malware development
2. anything security tooling related
3. reliable assembly
4. reliable and somewhat more obscure programming languages like blockchain programming langs or distributed programming with elixir
I feel that https://maldevacademy.com has been keeping my skills relatively fresh, it's fun in the discord where sometimes we try using models to do this stuff and it just chokes
On the other hand I "vibe coded" (Although I've heard more nuanced distinctions in terminology here about doing it right rather than just letting it rip automatically or without much thought) a fairly large and profitable software project in about a month recently that would have probably taken me over three if I hadn't used claude.
In the end and as you mentioned, these things are just tools and if you're busy complaining about them rather than finding what they can't do yet and having a bit of fun with working within those limits, you might be able to find a good niche to make better use of your time and avoid the skill atrophy
For me as well, I do projects for clients that would have taken me months, and I deliver them in a fraction of the time. It's amazing and totally helped cure my burnout (which I had for over a year before I found Claude Code.)
Speak for yourself. If anything is leading _me_ to burnout, it's this bullshit hype cycle.
I can instrument to observe the instructions being executed.
* Delegation
* Execution (better rules than rtk)
* Response
* Logging
When I want to turn off the printf style confirmations, manually editing saves the tokens. Using RFC 2119 (language) doesn't make a substantial difference. I'm in search of something better.
This is a reaction to the title, by someone who did not read the article.