The swindle goes like this, AI on a good codebase can build a lot of features, you think it’s faster it even seems safer and more accurate on times, especially in domains you don’t know everything about.
This goes in for a while whilst the codebase gets bigger and exploration takes longer and failure rate increases. You don’t want it to be true and try harder so you only stop after it practically became impossible to make any changes.
You look at the code again and there is so much code spaghetti is an understatement it’s the Chinese wall.
You start working…, and you realize what was going on
I deleted 75,000 of 140,000 lines of code and I honestly feel like the 3 months I went hard into agentic coding I wasted and I failed my users by building useless features increasing bugs, losing the mental model of my code and not finding the problems I didn’t know about the kind of hard decisions you only see when you in the code, the stuff that wanders in your mind for days
Coincidentally I've been working on a project for about 7 months now: its a 3d MMO. Currently its playable, and people are having fun with it - it has decent (but needs work) graphics, and you can cram a few hundred people into the server easily currently. The architecture is pretty nice, and its easy to extend and add features onto. Overall, I'm very happy with the progress, and its on track to launch after probably a years worth of development
In 7 months vibe coding, OP failed to produce a basic TUI. Maybe the feature velocity feels high, but this seems unbelievably slow for building a basic piece of UI like this - this is the kind of thing you could knock out in a few weeks by hand. There are tonnes of TUI libraries that are high quality at this point, and all you need to do is populate some tables with whatever data you're looking for. Its surprising that its taking so long
There seems to be a strong bias where using AI feels like you're making a lot of progress very quickly, but compared to manual coding it often seems to be significantly slower in practice. This seems to be backed up by the available productivity data, where AI users feel faster but produce less
This metric highly depends on who uses the AI to do what, where strong emphasis is on "who" and "what".
In my line of work (software developer) the biggest time sinks are meetings where people need to align proposed solutions with the expectations of stakeholders. From that aspect AI won't help much, or at all, so measuring the difference of man hours spent from solution proposal to when it ends up in the test loops with and without AI would yield... very disappointing results.
But for troubleshooting and fixing bugs, or actually implementing solutions once they have been approved? For me, I'm at least 10x'ing myself compared to before I was using AI. Not only in pure time, but also in my ability to reason around observed behaviors and investigating what those observations mean when troubleshooting.
But I also work with people who simply cannot make the AI produce valuable (correct) results. I think if you know exactly what you want and how you want it, AI is a great help. You just tell it to do what you would have done anyway, and it does it quicker than you could. But if you don't know exactly what you want, AI will be outright harmful to your progress.
AI, and especially agentic AI can make you lose situational awareness over a codebase and when you're doing deep work that SUUUUCKS, but it's not useless, you just have to play to it's strengths. Though my favorite hill to die on is telling people not to underestimate it's value as autocomplete. Turns out 40 gigabytes of autocomplete makes for a fucking amazing autocomplete. Try it with llama.vim + qwen coder 30b, it feels like the editor is reading your mind sometimes and the latency is so low.
1. If I use a coding agent to generate code, it should be something I am absolutely confident I can code correctly myself given the time (gun to my head test).
2. If it isn't, I can't move on until I completely understand what it is that has been generated, such that I would be able to recreate it myself.
3. I can create debt (I believe this is being called Cognitive Debt) by breaking rule 2, but it must be paid in full for me to declare a project complete.
Accumulating debt increases the chances that code I generate afterwards is of lower quality, and it also feels like the debt is compounding.
I'm also not really sure how these rules scale to serious projects. So far I've only been applying these to my personal projects. It's been a real joy to use agents this way though. I've been learning a lot, and I end up with a codebase that I understand to a comfortable level.
In their mind they’ve already done the “architectural heavy lifting” and accelerated the team. More often than not it just adds cognitive overhead where you spend more time deciphering and cleaning up garbage than actually building the thing properly from scratch.
But we don’t follow the same things for dependencies, work of colleagues, external services, all the layers down to the silicon when trying to work.
Why is AI suddenly different?
We just have to do this by risk and reward. What’s the downside if it’s wrong, and how likely is an error to be found in testing and review? What is the benefit gained if it’s all fine? This is the same for libraries and external services.
A complex financial set of rules in a non-updatable crypto contract with no testing?
A viewer for your internal log data to visualise something?
There are some programmers who treat the job as just plumbing together what is to them completely incomprehensible black boxes, who treat the computer as a mystery machine that just does things "somehow", but these programmers will almost always be hacks that spend their entire career producing mediocre code.
There are things such a programmer can build, but they are very limited by their lack of in depth understanding, and it is only a tiny fraction of what a more competent programmer can put together.
To get beyond being a hack, you need to understand the entire stack, including the code that you didn't write, including both libraries, frameworks and the OS, and including the hardware, the networking layers, and so forth. You don't have to be an expert at these things by any means, but you do need to understand them and be comfortable treating them as transparent boxes that you may have to go in and fiddle with at some points.
Comprehension debt just sounds like there are things you don’t (yet) understand.
Cognition debt means your lack of understanding compounds and the cognition “space” required to clear it increases accordingly.
An increasing comprehension debt that can be paid off one bit at a time within reasonable cognition space takes linear time to clear.
Cognition debt takes exponential time to clear the more of it you have. If it reaches a point where you simply don’t have the space for the cognition overhead required to understand the problem, you probably need to start over from your specifications.
Unfortunately, it is not, and many of its attempts at mathematical proofs have major flaws. You shouldn't trust its proofs unless you are already able to evaluate them--which I think is pretty much all the OP is saying.
There is one exception to this: If the AI also delivers the proof of why the math is correct, in a machine-checked format, and I understand the correctness theorem (not necessarily its proof). Then I would use it without hesitation.
I struggle to remember even relatively simple maths like working out "what percentage of X is Y" so if I write a formula like that I'll put in some simple values like 12 and 6 or 10,000 and 2,456 just to confirm I haven't got the values backwards or something. I've been shown sheets where someone put a formula in that they don't understand, checked it with numbers they can't easily eyeball and just assumed it was right as it's roughly in their ball park / they had no idea what the end result should be.
Then again I've also seen sheets where a 10% discount column always had a larger number than the standard price so even obviously wrong things aren't always checked.
I've reached solutions by trial and error too, and tried to rationalize them later, quite a few times. And it's easier to rationalize a working solution, however adversarial you claim to be in your rationalization.
I don't see using gen AI for the (not so) “brute force” exploration of the solution space as that different from trial and error and post fact rationalization.
Normally with mathematical problems you have to prove the solution correct. Testing is not sufficient, unless you can test all possible inputs exhaustively.
If it’s beyond our ability to review and we blindly trust it’s correct based on a limited set of tests… we’re asking for trouble.
I’m going to guess that this is Gell-Mann amnesia more than anything, and it’s going to get a lot of organizations into a lot of weird places.
"PhD level" just means you finished a bachelor and masters degree and are now doing a bit of original research as an employed research assistant.
Claude isn't "PhD level" anything. This shows a complete lack of understanding here. Claude has read every single text book in existence, so it can surface knowledge locked away in book chapters that people haven't read in years (nobody really reads those dense books on niche topics from start to finish).
Since Claude has infinite patience, you can just keep asking until you get it.
Your manager is unknowingly helping you create a form of job security for yourself, with all the technical debt and bugs being accumulated.
He might not understand it, and it might not be the type of work you want to do, but someone is going to have to fix those issues. And the longer they wait, the bigger the task gets.
Had a project idea which I coded with the help of AI and it became quite large to a point I was starting to have uncharted areas in the code. Mostly because I reviewed it too shallow or moved fast.
It was a good thing as that project never floated but if I were to do such a thing on my breadwinning project I would lose the joy.
This all works pretty great. Where it starts going off the rails is if I let it use a library I'm not >=90% comfortable with. That's a good use of these tools, but if I let it plow through feature requests, I end up accumulating debt, as you pointed out.
For my uses, I'm still finding the right balance. I'm not terribly sure it makes me faster. What I do think it helps with is longer focused sections because my cognitive load is being reduced. So I can get more done but not necessarily faster in the traditional sense. It's more that I can keep up momentum easier, which does deliver more over time.
I'm interested in multi agent systems, but I'm still not sure of the right orchestration pattern. These AI tools still can go off the rails real quick.
That’s the hard part of coding. If you have an architecture then writing the code is dead simple. If you aren’t writing the code you aren’t going to notice when you architected an API that allows nulls but then your database doesn’t. Or that it does allow that but you realize some other small issue you never accounted for.
I do not know how you can write this article and not realize the problem is the AI. Not that you let it architect, but that you weren’t paying attention to every single thing it does. It’s a glorified code generator. You need to be checking every thing it does.
The hard part of software engineering was never writing code. Junior devs know how to write code. The hard part is everything else.
The developers that thing coding is hard are the ones that absolutely love AI coding. It's changed their world because things they used to find hard are now easy.
Those that think coding is easy don't have such an easy time because coding to them is all about the abstractions, the maintainability and extensibility. They want to lay sensible foundations to allow the software to scale. This is the hard part. When you discover the right abstractions everything becomes relatively easy. But getting there is the hard part. These people find AI coding a useful tool but not the crazy amazing magical tool the people who struggle with coding do.
The OP is definitely in the second camp since they could spot and realise the shortcomings of the AI. They spotted the problem, and that problem is that the AI can't do the hard bit.
PMs can now cross reference and organize tickets with just a few keystrokes. Organisational knowledge, business knowledge, design systems and patterns, etc all of it is encoded in LLM consumable artefacts. For PMs it is the same switch - instead of having to do it by hand you direct lower level employees to handle the details and inconsistencies and you just do vibe and vision.
When all of the pieces successfully connect and execute reliably, what is left for humans to do? Just direct and consume?
And AI companies with their huge swaths of data are soon gonna be in the situation of being able to do the directing themselves
The first group are still thinking fairly deeply about design and interfaces and data structures, and are doing fairly heavy review in those areas. The second group are not, and those are the ones that I find a bit more worrisome.
I can't speak for others, but I'd go further and say that LLMs allow me to go deeper on the design side. I can survey alternative data structures, brainstorm conversationally, play design golf, work out a consistent domain taxonomy and from there function, data structure and field names, draft and redraft code, and then rewrite or edit the code myself when the AI cost/benefit trade off breaks down.
I’m not making a judgement call about which is better, but it was widely accepted in tech before the advent of LLMs that you just fundamentally lack a sense of understanding as a reviewer vs an author. It was a meme that engineers would rather just rewrite a complicated feature than fix a bug, because understanding someone else’s code was too much effort.
It's the same thing here. AI has dropped the cost of software development, so developers are now fooling themselves into producing low or zero value software. Since the value of the software is zero or near zero, it doesn't really matter whether you get it right or not. This freedom from external constraints lets you crank up development velocity, which makes you feel super productive, while effectively accomplishing less than if you had to actually pay a meaningful cost to develop something.
Like, what is the purpose of Gas Town? It looks to me like the purpose of Gas Town is to build Gas Town.
I find it useful to not listen to people who just talk.
I worry about the first group too, because interfaces and data structures are the map, not the territory. When you create a glossary, it is to compose a message, that transmit a specific idea. I find invariably that people that focus on code that much often forgot the main purpose of the program in favor of small features (the ticket). And that has accelerated with LLM tooling.
I believe most of us that are not so keen on AI tooling are always thinking about the program first, then the various parts, then the code. If you focus on a specific part, you make sure that you have well defined contracts to the orther parts that guarantees the correctness of the whole. If you need to change the contract, you change it with regard to the whole thing, not the specific part.
The issue with most LLM tools is that they’re linear. They can follow patterns well, and agents can have feedback loop that correct it. But contracts are multi dimensional forces that shapes a solution. That solution appears more like a collapsing wave function than a linear prediction.
I follow the plan -> red/green/refactor approach and it is surprisingly good, and the plans it produces all look super well reasoned and grounded, because the agent will slurp all the docs and forums with discussions and the like.
Trouble is once it starts working there would inevitably be a point where the docs and the implementation actually differ - either some combination of tools that have not been used in that way, some outdated docs, or just plain old bugs.
But if the goals of the project/feature are stated clearly enough it is quite capable of iterating itself out of an architectural dead end, that is if it can run and test itself locally.
It goes as deep as inspecting the code of dependencies and libraries and suggesting upstream fixes etc. all things that I would personally do in a deep debugging session.
And I’m supper happy with that approach as I’m more directing and supervising rather than doing the drudgery of it.
Trouble is a lot of my team mates _dont_ actually go this deep when addressing architectural problems, their usual mode of operandi is “escalate to the architect”.
This will not end up good for them in the long run I feel, but not sure what they can do themselves - the window of being able to run and understand everything seems to be rapidly closing.
Maybe that’s not super bad - I don’t exactly what the compiler is doing to translate things to machine code, and I definitely don’t get how the assembly itself is executed to produce the results I want at scale - that is level of magic and wizardry I can only admire (look ahead branching strategies and caching on modern cpus is super impressive - like how is all of this even producing correct responses reliable at such a a scale …)
Anyway - maybe all of this is ok - we will build new tools and frameworks to deal with all of this, human ingenuity and desire for improvement, measured in likes, references or money will still be there.
You need to be checking every thing it does.
This is also why I don't see it getting better anytime soon. So many people ask me "how do you get your claude to have such good output?" and the answer is always "I paid attention and spotted problems and asked claude to fix them." And it's literally that simple but I can see their eyes already glazing over.
Just as google made finding information easier, it didn't fix the human element of deciphering quality information from poor information.
You can skip that and go directly to writing code. But that meant you replaced a few hours of planning with a few weeks of coding.
> back to writing code by hand
But what they are doing is
> doing the __design work__ myself, by hand, before any code gets written.
So... Claude still is generating the code I guess?
And seriously, I can't understand that they thought their vibe coded project works fine and even bought a domain for the project without ever looking at source code it generated, FOR 7 MONTHS??
And the goal of the article is to draw attention to their project.
> Claude (c) by Anthropic (R) is the best thing since sliced bread and I'm Lovin' It(tm)! Here's a breakdown of you too can live a code free life for 10 easy payments of $99.99 a month if you subscribe now!
> Step one in your journey to code free life: code the whole damn project and put it together yourself
It's so much fluff and baloney and every single article is identical. And every single one is just over the top praise of Claude that doesn't come off as remotely authentic. There's always mentions of Claude "one shotting"(tm) something.
I don’t think it’s that weird to not look at the code if it’s a side project and you follow along incrementally via diffs. It’s definitely a different way of working but it’s not that crazy.
Its not weird to not look at the code, as long as you're looking at the code? (diffs?)
Uh, ok
We’ve moved to seeing that specs are useful and that having someone write lots of wrong code doesn’t make the project move faster (lots of times devs get annoyed at meetings and discussions because it hinders the code writing, but often those are there to stop everyone writing more of the wrong thing)
We’ve seen people find out that task management is useful.
Now more I’m seeing talk of fully doing the design work upfront. And we head towards waterfall style dev.
Then we’ll see someone start naming the process of prototyping, then I’m sure something about incremental features where you have to ma age old vs new requirements. Then talk of how really the customer needs to be involved more.
Genuinely, look at what projects and product managers do. They have been guiding projects where the product is code yet they are not expected to read the code and are required to use only natural language to achieve this.
When asking for a new major feature, despite hard guidelines and context (that eat half your context window), then it quickly ships bloat. The foundations are not very well organized and this is where you acknowledge it is all about random-prediction of the next word-thing.
Overall, i've wasted more time reviewing the PR and trying to steer it properly than I expected. So multi-layer agent vibe coding is no longer the way to go *for me*. Maybe with unlimited tokens and a better prompt, to be investigated...
This is a special case of a general fundamental point I'm struggling with.
Let's assume AI has reduced the marginal cost of code to zero. So our supply of code is now infinite.
Meanwhile, other critical factors continue to be finite: time in a day, attention, interest, goodwill, paying customers, money, energy.
So how do you choose what to build?
Like a genie, the tools give us the power to ask for whatever we want. And like a genie, it turns out we often don't really know what we want.
Now it is different in a way where now I don’t have time to use those apps.
That’s a joke.
But I do believe it answers the question of “what to build?”. If you didn’t have time before LLM assisted coding you still don’t have time for it. You most likely know what is used and what not already by heart or by some measurements.
The very worst things you can do in a codebase are (a) not deeply understand how it works (have it be magic) and (b) be lazy and mess up the structure.
How do you fix a problem which happens at 2:00am and takes your system down if you don't have an excellent understanding of how it works?
Over time we're already bad at (a) because most developers hate writing documentation so that knowledge is invariably lost over time.
And I'm sure the rewrite is going to teach me a whole different set of lessons...
Not sure why good coverage wouldn't mitigate risk in a refactor...
My mantra whenever I'm working with AI is that I want it to know what "point b" looks like and be able to tell by itself whether it's gotten there...
If you have a working implementation, it sounds like you have a basis for automated tests to be written... once you have that (assuming that the tests are written to test the interface rather than the implementation), then it should be fairly direct to have an agent extract and decompose...
For example, consider a lint rule that bans Kysely queries on certain tables from existing outside of a specific folder. You'd write a rule like this in an effort to pull reads and writes on a certain domain into one place, hoping you can just hand the lint violations to your AI agent and it would split your queries into service calls as needed.
And at first, it will appear to have Just Worked™. You are feeling the AGI. Right up until you start to review the output carefully. Because there are now little discrepancies in the new queries written (like not distinguishing between calls to the primary vs. the replica, missing the point of a certain LIMIT or ORDER BY clause, failing to appropriately rewrite a condition or SELECT, etc.) You run a few more reviewer agent passes over it, but realize your efforts are entirely in vain... because even if the reviewer agent fixes 10 or 20 or 30 of the issues, you can still never fully trust the output.
As someone with experience in doing this kind of thing before AI, I went back to doing it the old way: using a codemod to rewrite the code automatically using a series of rules. AI can write the codemod, AI can help me evaluate the results, but actually having it apply all of the few hundred changes automatically led to a lack of my ability to trust the output. And I suspect that will continue to be true for some time.
This industry needs a "verification layer" that, as far as I know, it does not have yet. Some part of me hopes that someone will reply to this comment with a counterexample, because I could sorely use one.
A really screwed code base blows out your context window and just starts burning tokens as the AI works out a way to kill -9 itself to escape the hell you're subjecting it to.
The rewrite is me sitting down with a blank doc and drawing the boxes before any code exists. Then the CLAUDE.md enforces what I already decided. Whether that actually holds up as the project grows, I genuinely don't know yet.
I have found small iterations to have the best results. I'm not giving AI any chance to one shot it. For example, I won't tell it to "create a fleet view" but something more like "extract key binding to a service" so that I can reuse it in another view before adding another view. Basically, talk to the AI as an engineer talking to another engineer at the nitty gritty level that we need to deal with everyday, not a product person wishing for a business selling point to magically happen.
Personally, I've taken the time its freed up to spend more time on mathacademy and reading more theory oriented books on data structures and algorithms. AI coding systems are at their best when paired with someone with broad knowledge. knowing what to ask for and knowing the vocabulary to be specific about what you want to be built is going to be a much more valuable job skill going forward.
One example is a small AI based learning system I have been developing in my free time to help me learn. the mvp stored an entire knowledge graph and progress in markdown files. being an engineer, I knew this wouldn't scale so once I proved the concept viable, I moved everything into sqlite with a graphdb. then I decided to wrap some parts of teh functionality in to rust and put everything behind a small rust layer with the progress tracking logic still being in python.
someone with no knowlege of graph databases or dependncy graphs or heuristics would not be able to build this even if they had AI. they simply don't know what they dont' know and AI wont' save you there.
That said, I think its important to also spend time in the dirt. I've recently started pickign up zig as my NO AI langauge just to keep. those skills sharp.
I'm really curious if we'll seesaw once AI costs go up 10x.
Eventually like every hype wave the dust will settle, and lets see where we stand.
By now all the AI companies have consumed all human knowledge so they either learn to actually think for themselves, or that is it.
Either way, that won't change the ongoing layoffs while trying to pursue the AI dream from management point of view.
I think most companies doing layoffs are bloated to begin with, AI is just the scapegoat to do the layoffs.
Translation and asset generation teams for enterprise CMS, whose role has now been taken by AI.
Likewise traditional backend development, that was already reduced via SaaS products, serverless, iPaaS low code/no code tooling, that now is further reduced via agents workflow tooling, doing orchestration via tools (serverless endpoints).
Isn't Golang relatively easier to read than Rust? I was under the impression that Rust is a more complex language syntactically.
> The other change is simpler: I'm doing the design work myself, by hand, before any code gets written. Not a vague doc. Concrete interfaces, message types, ownership rules. The architecture decisions that the AI kept making wrong are now made in writing before the first prompt.
This post is good to grasp the difference between "vibe-coding" and using the AI to help with design and architectural choices done by a competent programmer (I am not saying you are not one). Lately I feel that Opus 4.7 involves the user a lot more, even when given a prompt to one-shot a particular piece of software.
+1 on Open 4.7 involving the user a lot more. Rn I'm trying to get to a state where I can codify my design + decision preferences as agents personas and push myself out of the dev loop.
Good architecture in any language is obvious to someone who is experienced and cares.
Go is actually great for bots to write if you’re actually thinking.
> Go reads fine whether the architecture is good or bad
Were you reading the Golang code all along and got fooled or did you review it after it failed? Sorry I admit I didn't read the whole article.
It sounds like the author knows Rust, and might not be as familiar with Go.
A language that you are proficient in is always going to be easier read than one you don’t, even if it is an objectively easier language to to read in general.
I’ve used AI tools to do i18n translations to Spanish and Portuguese (somewhat ashamed to admit this). I’ve grown more familiar with the structure of these languages, and come to recognize some of the common vocabulary for our agtech domain. If anything, I feel more clueless about both languages now than I did before, when it comes to any sort of proficiency.
Hey I don't want to over simplify, I'm sure it was complicated, but did the author have functional tests for these broken views? As long as there are functional tests passing on the previous commit I'd have thought that claude could look at the end situation and work out how to get the desired feature without breaking the other stuff.
TUIs aren't an exception, it's still essential to have a way to end-to-end test each view.
You can't test every permutation of app usage. You actually need good architechture so you can trust your test and changes to be local with minimal side-effects.
I still do, but I used to, too.
But I will say... you have to know Golang. You have to have at least tried to make a BubbleTea app yourself and try to understand ELM architecture. You have to look at the code and increment with it.
It makes total sense for OP to switch to Rust and Ratatui if they don't know Golang well. But I don't think it's a better language for it. [Ratatui has brought me great inspiration though!]
Independent of framework, the LLMs get the spacial relationships. I say things like "the upper right panel's content is not wrapping inside and the panel's right edge should extend to the terminal edge" and the LLM will fix it. They can see the resultant text, I'm copy-pasting all the time.
TUI code is finicky; one mis-rendered component mucks everything up. The LLMs will decide themselves make little, temporary BubbleTea fixtures to help understand for itself when things aren't right.
The only real problem with LLMs and BubbleTea is that upon first prompt, they insist on using BubbleaTea v1 versus BubbleTea v2, released in December 2025. But then you just point it to the V2_UPGRADE.md and it gets back on track. That will improve as training cutoffs expand.
I vibe-coded this TUI for Mom's last night. I actually started with Grok (who started with v1) and then moved into Claude Code after some iteration:
https://gist.github.com/neomantra/1008e7f2ad5119d3dd5716d52e...
Also 1600 lines... didn't any agent reviewing the diffs point that out?
You're also adding a lot to claude.md, I dunno how much that file has grown but a big claude.md file with many instructions, I don't think the ai will be able to remember all those rules.
In my experience, no. These tools suck at refactoring, mostly choosing to add more code instead.
Also 1600 lines... didn't any agent reviewing the diffs point that out?
You're also adding a lot to claude.md, I dunno how much that file has grown but a big claude.md file with many instructions, I don't think the ai will be able to remember all those rules
The ones who are “AI pilled” and the contagious lepers.
Do they write empty functions and let AI fill them in?
Or do they use some kind of specification language?
Are people designing those languages?
That trial and error process is still happening with a LLM, but much faster, and with instantaneous cross-references to various forms of documentation that I would be looking up myself otherwise. It produces code of a quality that is dependent on the engineer knowing what they want in the first place and prompting for it and refining its output correctly.
It's the exact same process of sculpting code that the majority of the industry was doing "by hand" prior to the release of LLMs, but faster, and the harnesses are only getting better. To "vibe code" is to prompt vaguely and ignore the quality of the output. You're coming to a forum full of professionals and essentially telling us that you're getting really frustrated with your Scratch project.
I don't know if you're trying to lead a charge or whatever but good luck with that. As a senior SWE, it is clear to me that this is the new paradigm until something better than LLMs comes along. My workflows and efficiency have been vastly improved. I will admit that I have never really been a "I made a SMTP server in 3k of Rust" kind of guy, though.
If you understand good software architecture, architect it. Create a markdown document just as you would if you had a team of engineers working with you and would hand off to them. Be specific.
Let the AI do the implementation of your architecture.
This. I definitely agree with this statement at this point in AI-assisted development. This gets at the "taste" factor that is still intrinsically human, especially in software engineering. If you can construct and guide the overall architecture of an application or system, AI can conceivably fill in the smaller feature bits, and do so well. But it must have a strong architecture and opinionated field in which to play.
With that said, this caught my eye:
> AI gravitates toward single-struct-holds-everything because it satisfies the immediate prompt with minimal ceremony.
This is too general. "AI" is used here as a catch-all, but in fact, it was the specific model under the specific conditions you ran your prompt, including harness, markdowns, PRDs, etc. So it's not fair to say "AI does X!" in this case.
It's also very much up to you. It's very common to have a frontier model plan an architecture before you have another model implement code. If you're just one-shotting an LLM to do everything you get mediocre, more brittle code.
This stuff is still being figured out by a lot of people. But I feel the core of the issue is not using AI well. Scoping, task alignment, validation, are crucial.
some states, for an example, are meant to be assumed from the data shape, rather than the actual state fields, but damn they like adding a state field.
It would have been easy to run a few ai agents to review the code and find these issues as well and architect it clearly
Yea, that's why engineers are still very important for now (until models can do this type of longer term designs and stick to them).
This is what I was doing right from the beginning. AI just fills out methods and doing other low intelligence work. Both are happy. My architectures and code are really mine, easy to read and reason. AI gets paid and does not get a chance to fuck me in the process. At no point I felt any temptation to leave "serious" to AI.
Inb4 “you’re gonna be replaced” god damn it I hope so, I do not want to spend the rest of my life behind a computer screen…
But here's the thing, you almost never know what the architecture is up front. If you do you probably aren't the one writing the actual code anymore. Writing the code, with or without an AI is part of the design process. For most people it isn't until they've tried several times, fucked it up a bunch, and refactored or rewrote even more that you actually know what the architecture needs to be.
7 months ago was early November. Coding assistants were getting very good back then, but they were still significantly poorer at making good architectural decisions in my experience. They tended to just force features into the existing code base without much thought or care.
Today I've noticed assistants tend to spot architectural smells while working and will ask you whether they should try to address it, but even then they're probably never going to suggest a full refactor of the codebase (which probably is generally the correct heuristic).
My guess is that if you built this today with AI that you wouldn't run into so many of these problems. That's not to say you should build blind, but the first thing that stood out to me was that you starting building 7 months ago and coding assistants were only just becoming decent at that time, and undirected would still generally generate total slop.
Now I do feel lucky that I started learning coding about four years before the LLM revolution, but these things are really just natural language compilers, aren’t they? We’re just in that period - the 1980s, the greybeards tell me - where companies charged thousands of dollars per compiler instance, right? And now, I myself have never paid for a compiler.
This whole investor bubble will blow up in the face of the rentier-finance capitalists and I’ll be laughing my head off while it happens.
Time to become a "product engineer" and watch the hyper-agile agents putting up digital post-it notes on digital pin-boards discussing how much each post-it is worth in digital scrum meetings. Meanwhile the agents keep wasting more and more time so that their owners make less and less of a lose, until eventually a profit is made.
Until the costs become prohibitive and humans become cheaper than the agents that replaced them. Once the agents are replaced by the humans, the next hype bubble awaits around the bend.
/s