Get the top HN stories in your inbox every day.
viccis
Haven't seen this mentioned yet, but the worst part for me is that a lot of management LOVES to use Claude to generate 50 page design documents, PRDs, etc., and send them to us to "please review as soon as you can". Nobody reads it, not even the people making it. I'm watching some employees just generate endless slide decks of nonsense and then waffle when asked any specific questions. If any of that is read, it is by other peoples' Claude.
It has also enabled a few people to write code or plan out implementation details who haven't done so in a long (sometimes decade or more) time, and so I'm getting some bizarre suggestions.
Otherwise, it really does depend on what kind of code. I hand write prod code, and the only thing that AI can do is review it and point out bugs to me. But for other things, like a throwaway script to generate a bunch of data for load testing? Sure, why not.
jermaustin1
I've been tasked with code reviews of Claude chat bot written code (not Claude code that has RAG and can browse the file system). It always lacks any understanding of our problem area, 75% of the time it only works for a specific scenario (the prompted case), and almost 100% of the time, when I comment about this, I'm told to take it over and make it work... and to use Claude.
I've kind of decided this is my last job, so when this company folds or fires me, I'm just going to retire to my cabin in the rural Louisiana woods, and my wife will be the breadwinner. I only have a few 10s of thousands left to make that home "free" (pay off the mortgage, add solar and batteries, plant more than just potatoes and tomatoes).
Though, post retirement, I will support my wife's therapy practice, and I have a goal of silly businesses that are just fun to do (until they arent), like my potato/tomato hybrid (actually just a graft) so you can make fries and ketchup from the same plant!
windex
>so you can make fries and ketchup from the same plant!
We should be friends. I like your ideas.
jermaustin1
I'm always looking for people to share my weird ideas that have absolutely nothing to do with software or computers. Unfortunately my only friends are all software people who have no interest outside of computers. Something I've found to have very little interest in anymore.
Zecc
For what is worth, I prefer the name pomato to totato.
jermaustin1
I'll keep that in mind when marketing. I was going to go with French Fry Tree.
jasondigitized
You got any land down there? I would like to be close to you and, post retirement, eat said french fries daily.
jermaustin1
This might be a little dark, but the majority of our street is very elder, and none of there families want to move over here.
They were the original non-familial homesteaders from 50+ years ago when all this land was my wife's great grandfather's, and he sold off small plots to people. He, infact, inherited it from his father, who bought a half mile square back in the 20s or 30s (I believe). The first house on the road was his (Great Great Grandpa). The road WAS his driveway, then slowly but surely new generations of the family started building houses a few hundred yards away from each other, then they started selling plots to people in the 60s, and sold the last of the original land in 2023 about a year before grandpa passed.
Now the only land left in "the family", is this 1.25 acre plot that I live on. I don't really have the desire to buy more from the folks that are dying, but my neighbor has already bought up about half of the vacant land.
tbreschi
Best comment of the day
0_____0
That sounds lovely. I think too many people get attached to the structure of life as they've lived it for the last n years and resist natural phase transitions for far too long. Good luck with retirement and your dream of being the botanical equivalent of the mean kid from Toy Story:p
BoneShard
I noticied what previously would take 30 mins, now takes a week. For example we had a performance issue with a DB, previously I'd just create a GSI (global secondary index), now there is a 37 page document with explanation, mitigation, planning, steps, reviews, risks, deployment plan, obstacles and a bunch of comments, but sure it looks cool and very professional.
ionwake
Im now out of the workforce and can’t even imagine the complexity of the systems as management and everyone else communicate plans and executions through Claude. It must already be the case that some code based are massive behemoths few devs understand. Is Claude good enough to help maintain and help devs stay on top of the codebase?
BoneShard
The code is fine, strong reviews help and since we're slower due to all slop communication also helps.
floydian10
Not to mention that now you need an "alignment meeting" with a bunch of people
maxaw
I quit my last job because of this. I’m pretty sure manager was using free chatgpt with no regard for context length too, because not only was it verbose it was also close to gibberish. Being asked to review urgently and estimate deadlines got old real fast
theshrike79
If you shove clearly AI generated content at me, I will use an AI to summarize it.
Or I'll walk up to your desk and ask you to explain it.
DalekBaldwin
Jump straight to the second option. You have to presume that the content they sent you has no relation whatsoever to their actual understanding of the matter.
vdfs
Be prepared for "I Asked claude and it said: ..." at some point you will just ask claude via a microphone
Arubis
I actually think there’s almost an acceptable workflow here of using LLMs as part of the medium of communication. I’m pretty much fine with someone sending me 500 lines of slop with the stated expectation that I’ll dump it into an LLM on my end and interact with it.
It’s the asymmetric expectations—that one person can spew slop but the other must go full-effort—that for me personally feels disrespectful.
SOLAR_FIELDS
I also don't mind that. Summarized information exchange feels very efficient. But for sure, it seems like a societal expectation is emerging around these tools right now - expect me to put as much effort into consuming data as you did producing it. If you shat out a bunch of data from an LLM, I'm going to use an LLM to consume that data as well. And it's not reasonable for you to expect me to manually parse that data, just as well as I wouldn't expect you to do the same.
However, since people are not going to readily reveal that they used an LLM to produce said output, it seems like the most logical way to do this is just always use an LLM to consume inputs, because there's no easy 100% way to tell whether it was created by an LLM or a human or not anymore.
arppacket
I think we'll eventually move away from using these verbose documents, presentations, etc for communication. Just do your work, thinking, solving problems, etc while verbally dumping it all out into LLM sessions as you go. When someone needs to be updated on a particular task or project, there will be a way to give them granular access to those sessions as a sort of partial "brain dump" of yours. They can ask the LLM questions directly, get bullet points, whatever form they prefer the information in.
That way, thinking is communication! That's kind of why I loved math so much - it felt like I could solve a problem and succinctly communicate with the reader at the same time.
etothepii
If you write 3 bullet points and produce 500-pages of slop why would my AI summarise it back to the original 3 bullet points and not something else entirely?
0x262d
is this better than normal communication in any way, or just not much worse?
rdevilla
> It’s the asymmetric expectations—that one person can spew slop but the other must go full-effort—that for me personally feels disrespectful.
This has always been the case. Have some junior shit out a few thousand lines of code, leave, and leave it for the senior cleanup crew to figure out what the fuck just happened...
hoosieree
"send me your prompts instead"
theshrike79
There's a discussion going on that if you use an LLM to generate code, should the prompts (and related stuff) be a part of the pull request.
rkagerer
If you shove content at me that I even suspect was AI generated I will summarily hit the delete button and probably ban you from sending me any form of communication ever again.
It's a breach of trust. I don't care if you're my friend, my boss, a stranger, or my dog - it crosses a line.
I value my time and my attention. I will willingly spend it on humans, but I most certainly won't spend it on your slop when you didn't even feel me worth making a human effort.
__blockcipher__
I highly recommend you let your dog use LLMs. They have trouble composing long messages on human-centric keyboards.
michaelteter
Obviously you should also use Claude to consume those 50 pages. It sounds cynical, but it's not. It's practical.
What I've learned in 2 years of heavy LLM use - ChatGPT, Gemini, and Claude, is that the significance is on expressing and then refining goals and plans. The details are noise. The clear goals matter, and the plans are derived from those.
I regularly interrupt my tools to say, "Please document what you just said in ...". And I manage the document organization.
At any point I can start fresh with any AI tool and say, "read x, y, and z documents, and then let's discuss our plans". Although I find that with Gemini, despite saying, "let's discuss", it wants to go build stuff. The stop button is there for a reason.
LordGrey
I use an agents.md file to guide Claude, and I include a prominent line that reads UPDATE THIS FILE WITH NEW LEARNINGS. This is a bit noisy -- I have to edit what is added -- but works well and it serves as ongoing instruction. And as you have pointed out, the document serves as a great base if/when I have to switch tools.
MarsIronPI
I've found in my (admittedly limited) use of LLMs that they're great for writing code if I don't forsee a need to review it myself either, but if I'm going to be editing the code myself later I need to be the one writing it. Also LLMs are bad at design.
mcswell
Master Foo and the Programming Prodigy: https://catb.org/~esr/writings/unix-koans/prodigy.html
ignoramous
> Also LLMs are bad at design.
I've found that SoTA LLMs sometimes implement / design differently (in the sense that "why didn't I think of that"), and that's always refreshing to see. I may run the same prompt through Gemini, Sonnet, and Codex just to see if they'd come up with some technique I didn't even know to consider.
> don't forsee a need to review it myself either
On the flip side, SoTA LLMs are crazy good at code review and bug fixes. I always use "find and fix business logic errors, edge cases, and api / language misuse" prompt after every substantial commit.
0x6d61646f
what code do you write that you don't need to mantain/read again later?
tikotus
For me it's throwaway scripts and tools. Or tools in general. But only simple tools that it can somewhat one-shot. If I ever need to tweak it, I one-shot another tool. If it works, it's fine. No need to know how it works.
If I'm feeling brave, I let it write functions with very clear and well defined input/output, like a well established algorithm. I know it can one-shot those, or they can be easily tested.
But when doing something that I know will be further developed, maintained, I mainly end up writing it by hand. I used to have the LLM write that kind of code as well, but I found it to be slower in the long run.
tracker1
Definitely a lot of one-shot scripts for a given environment... I've started using a run/ directory for shell scripts that will do things like spin up a set of containers defined in a compose file.. build and test certain sub-projects, initialize a database, etc.
For the most part, many of them work the first time and just continue to do so to aid a project. I've done similar in terms of scaffolding a test/demo environment around a component that I'm directly focused on... sometimes similar for documentation site(s) for gh pages, etc.
Soem things have gone surprisingly well.
alper
One group of people pretends to have written something and another group of people pretends to have read something. Much productivity is gained.
Zizek had a great point about this.
ZoomZoomZoom
At least both get paid in not-pretend money.
andrei_says_
For the time being. Their manager is under constant pressure to lay them off and replace them with “ai”.
deadbabe
The best thing to do is to schedule meetings with those people to go over the docs with them. Now you force them to eat their own shit and waste their own time the more output they create.
caminante
Love the intent, but isn't that wishful if you don't have any leverage? e.g., the higher up will trade you for someone who doesn't cause friction or you waste too much of your own time?
ceejayoz
I had Claude review one. It was... not complimentary. Seemed to help a bit.
hdhdhsjsbdh
It has made my job an awful slog, and my personal projects move faster.
At work, the devs up the chain now do everything with AI – not just coding – then task me with cleaning it up. It is painful and time consuming, the code base is a mess. In one case I had to merge a feature from one team into the main code base, but the feature was AI coded so it did not obey the API design of the main project. It also included a ton of stuff you don’t need in the first pass - a ton of error checking and hand-rolled parsing, etc, that I had to spend over a week unrolling so that I could trim it down and redesign it to work in the main codebase. It was a slog, and it also made me look bad because it took me forever compared to the team who originally churned it out almost instantly. AI tools are not good at this kind of design deconflicting task, so while it’s easy to get the initial concept out the gate almost instantly, you can’t just magically fit it into the bigger codebase without facing the technical debt you’ve generated.
In my personal projects, I get to experience a bit of the fun I think others are having. You can very quickly build out new features, explore new ideas, etc. You have to be thoughtful about the design because the codebase can get messy and hard to build on. Often I design the APIs and then have Claude critique them and implement them.
I think the future is bleak for people in my spot professionally – not junior, but also not leading the team. I think the middle will be hollowed out and replaced with principals who set direction, coordinate, and execute. A privileged few will be hired and developed to become leaders eventually (or strike gold with their own projects), but everyone in between is in trouble.
ramraj07
If you dont take a stand and refuse to clean their mess, aren't you part of the problem? No self respecting proponent of AI enabled development should suggest that the engineers generating the code are still not personally responsible for its quality.
tenacious_tuna
Ultimately that's only an option if you can sustain the impact to your career (not getting promoted, or getting fired). My org (publicly traded, household name, <5k employees) is all-in on AI with the goal of having 100% of our code AI generated within the next year. We have all the same successes and failures as everyone else, there's nothing special about our case, but our technical leadership is fundamentally convinced that this is both viable and necessary, and will not be told otherwise.
People who disagree at all levels of seniority have been made to leave the organization.
Practically speaking, there's no sexy pitch you can make about doing quality grunt work. I've made that mistake virtually every time I've joined a company: I make performance improvements, I stabilize CI, I improve code readability, remove compiler warnings, you name it: but if you're not shipping features, if you're not driving the income needle, you have a much more difficult time framing your value to a non-engineering audience, who ultimately sign the paychecks.
Obviously this varies wildly by organization, but it's been true everywhere I've worked to varying degrees. Some companies (and bosses) are more self-aware than others, which can help for framing the conversation (and retaining one's sanity), but at the end of the day if I'm making a stand about how bad AI quality is, but my AI-using coworker has shipped six medium sized features, I'm not winning that argument.
It doesn't help that I think non-engineers view code quality as a technical boogeyman and an internal issue to their engineering divisions. Our technical leadership's attitude towards our incidents has been "just write better code," which... Well. I don't need to explain the ridiculousness of that statement in this forum, but it undermines most people's criticism of AI. Sure, it writes crap code and misses business requirements; but in the eyes of my product team? That's just dealing with engineers in general. It's not like they can tell the difference.
sjducb
Hi thanks for this brilliant feature. It will really improve the product. However it needs a little bit more work before we can merge it into our main product.
1) The new feature does not follow the existing API guidelines found here: see examples an and b.
2) The new feature does not use our existing input validation and security checking code, see example.
Once the following points have been addressed we will be happy to integrate it.
All the best.
The ball is now in their court and the feature should come back better
This is a politics problem. Engineers were sending each other crap long before AI.
zaphar
There is an alternative way make the necessary point here.. Let it go through with comments to the effect that you can not attest to the quality or efficacy of the code and let the organization suffer the consequences of this foray into LLM usage. If they can't use these tools responsibly and are unwilling to listen to the people who can, then they deserve to hit the inevitable quality wall Where endless passes through the AI still can't deliver working software and their token budget goes through the ceiling attempting to make it work.
diacritical
> My org [...] is all-in on AI with the goal of having 100% of our code AI generated within the next year.
> People who disagree at all levels of seniority have been made to leave the organization.
So either they're right (100% AI-generated code soon) and you'll be out of a job or they'll be wrong, but by then the smart people will have been gone for a while. Do you see a third future where next year you'll still have a job and the company will still have a future?
LordGrey
> ... I make performance improvements, I stabilize CI, I improve code readability, remove compiler warnings, you name it ...
These are exactly the kind of tasks that I ask an AI tool to perform.
Claude, Codex, et al are terrible at innovation. What they are good at is regurgitating patterns they've seen before, which often mean refactoring something into a more stable/common format. You can paste compiler warnings and errors into an agentic tool's input box and have it fix them for you, with a good chance for success.
I feel for your position within your org, but these tools are definitely shaking things up. Some tasks will be given over entirely to agentic tools.
mentalgear
Unfortunately not many companies seem to require engineers to cycle between "feature" and "maintainability" work - hence those looking for the low-hanging fruits and know how to virtue signal seem to build their career on "features" while engineers passionate about correct solutions are left to pay for it while also labelled as "inefficient" by management. It's all a clown show, especially now with vibe-coding - no wonder we have big companies having had multiple incidents since vibing started taking off.
whiplash451
Shipping “quality only” work for a long time can be stressful for your colleagues and the product teams.
You’re much better off mixing both (quality work and product features).
oh_my_goodness
"aren't you part of the problem?"
Yes? In the same way any victim of shoddy practices is "part of the problem"?
ramraj07
Employees, especially ones as well leveraged and overpaid as software engineers, are not victims. They can leave. They _should_ leave. Great engineers are still able to bet better paying jobs all the time.
borski
Came here to say this. The right solution to this is still the same as it always was - teach the juniors what good code looks like, and how to write it. Over time, they will learn to clean up the LLM’s messes on their own, improving both jobs.
Aurornis
> and refuse to clean their mess
You can should speak up when tasks are poorly defined, underestimated, or miscommunicated.
Try to flat out “refuse” assigned work and you’ll be swept away in the next round of layoffs, replaced by someone who knows how to communicate and behave diplomatically.
arwhatever
ramraj07 went on to clarify that they were advocating for putting the onus for cleanup back on mess generators.
They clearly were not advocating for flat out refusing.
theshrike79
Just reply with this to every AI programming task: https://simonwillison.net/2025/Dec/18/code-proven-to-work/
It's just plain unprofessional to just YOLO shit with AI and force actual humans to read to code even if the "author" hasn't read it.
Also API design etc. should be automatically checked by tooling and CI builds, and thus PR merges, should be denied until the checks pass.
dude250711
> It was a slog, and it also made me look bad because it took me forever compared to the team who originally churned it out almost instantly.
The hell you are playing hero for? Delegate the choice to manager: ruin the codebase or allocate two weeks for clean-up - their choice. If the magical AI team claim they can do integration faster - let them.
jcgrillo
IME one thing that makes this choice a very difficult one is oncall responsibilities. The thing that incentivizes code owners to keep their house in order is that their oncall experience will be a lot better. And you're the only one who is incentivized to think this way. Management certainly doesn't care. So by delegating the choice to management you're signing up for a whole bunch of extra work in the form of sleepless oncall shifts.
johntash
If someone is making the kind of mistakes that cause oncall issues to increase, put that person on call. It doesn't matter if they can't do anything, call them every time they cause someone else to be paged.
IME too many don't care about on call unless they are personally affected.
phyzix5761
> did not obey the API design of the main project
If they're handing you broken code call them out on it. Say this doesn't do what it says it does, did you want me to create a story for redoing all this work?
AnimalMuppet
I've heard of human engineers who are like that. "10x", but it doesn't actually work with the environment it needs to work in. But they sure got it to "feature complete" fast. The problem is, that's a long way from "actually done".
suzzer99
> At work, the devs up the chain now do everything with AI – not just coding – then task me with cleaning it up.
This has to be the most thankless job for the near future. It's hard and you get about as much credit as the worker who cleans up the job site after the contractors are done, even though you're actually fixing structural defects.
And god forbid you introduce a regression bug cleaning up some horrible redundant spaghetti code.
hdhdhsjsbdh
Near future being the key term here imo. The entire task I mentioned was not an engineering problem, but a communication issue. The two project owners could have just talked to each other about the design, then coded it correctly in the first pass, obviating the need for the code janitor. Once orgs adapt to this new workflow, they’ll replace the code janitors with much cheaper Claude credits.
j3k3
Lol you may be on to something there.. 'a code janitor'.
ehnto
Thst is definitely one tell, the hand rolled input parsing or error handling that people would never have done at their own discretion. The bigger issue is that we already do the error checking and parsing at the different points of abstraction where it makes the most sense. So it's bespoke, and redundant.
That is on the people using the AI and not cleaning up/thinking about it at all.
dawnerd
We’ve had this too and made a change to our code review guidelines to mention rejection if code is clearly just ai slop. We’ve let like four contractors go so far over it. Like ya they get work done fast but then when it comes to making it production ready they’re completely incapable. Last time we just merged it anyways to hit a budget it set everyone back and we’re still cleaning up the mess.
fastasucan
It makes my work suck, sadly. Team dynamics also contributes to that, admittedly.
Last year I was working on implementing a pretty big feature in our codebase, it required a lot of focus to get the business logic right and at the same time you had be very creative to make this feasible to run without hogging to much resources.
When I was nearly done and worked on catching bugs, team members grew tired of waiting and starting taking my code from x weeks ago (I have no idea why), feeding it to Claude or whatever and then came back with a solution. So instead of me finishing my code I had to go through their version of my code.
Each one of the proposals had one or more business requirements wrong and several huge bugs. Not one was any closer to a solution than mine was.
I had appreciated any contribution to my code, but thinking that it would be so easy to just take my code and finishing it by asking Claude was rather insulting.
boredemployee
I completely understand.
We're in a phase where founders are obsessed with productivity so everything seens to work just fine and as intended with few slops.
They're racing to be as productive as possible so we can get who knows where.
There are times when I honestly don't even know why we're automating certain tasks anymore.
In the past, we had the option of saying we didn't know something, especially when it was an area we didn't want to know about. Today, we no longer have that option, because knowledge is just a prompt away. So you end up doing front-end work for a backend application you just built, even though your role was supposed to be completely different.
xantronix
This feels similar to the slow encroachment of devops onto everything. We're making so much shit nowadays that there is nobody left but developers to shepherd things into production, with all the extra responsibility and none of the extra pay commensurate with being a sysadmin too.
bluefirebrand
> Today, we no longer have that option, because knowledge is just a prompt away
Something resembling knowledge anyway. A sort of shambling mound wearing knowledge like a skinsuit
boredemployee
While I agree, I can't deny that AI is doing the job most of the time. But the hunt for the supreme productivity feels disgusting sometimes.
2muchcoffeeman
There’s a lot more going on there than AI …
rsoto2
Not really, this is exactly what I expect due to baseless lies from the AI companies and a disdain for employee payroll by the C-suite.
foolserrandboy
they fantasize about unpaid interns writing specs and nobody ever needed to look at the code in a few years
undefined
peab
This seems to be a team problem more than anything? Why are your coworkers taking on your responsibilities? Where's your manager on this?
nsxwolf
Could be an emergent team problem that wouldn’t have had cause to exist before AI.
undefined
unsaved159
Not really AI problem, more like garbage coworkers.
Izkata
I don't use it.
I know my mind fairly well, and I know my style of laziness will result in atrophying skills. Better not to risk it.
One of my co-workers already admitted as much to me around six months ago, and that he was trying not to use AI for any code generation anymore, but it was really difficult to stop because it was so easy to reach for. Sounded kind of like a drug addiction to me. And I had the impression he only felt comfortable admitting it to me because I don't make it a secret that I don't use it.
Another co-worker did stop using it to generate code because (if I'm remembering right) he can tell what it generates is messy for long-term maintenance, even if it does work and even though he's new to React. He still uses it often for asking questions.
A third (this one a junior) seemed to get dumber over the past year, opening merge request that didn't solve the problem. In a couple of these cases my manager mentioned either seeing him use AI while they were pairing (and it looked good enough so the problems just slipped by) or saw hints in the merge request with how AI names or structures the code.
Xcelerate
I've been using ChatGPT to teach myself all sorts of interesting fields of mathematics that I've wanted to learn but never had the time previously. I use the Pro version to pull up as many actual literature references as I can.
I don't use it at all to program despite that being my day job for exactly the reason you mentioned. I know I'll totally forget how to program. During a tight crunch period, I might use it as a quick API reference, but certainly not to generate any code. (Absolutely not saying it's not useful for this purpose—I just know myself well enough to know how this is going to go haha)
minhaz23
How do you get chatgpt to teach you well? I feel like no matter how dense and detailed i ask it to be or how much i ask it to elaborate and contextualize topics with their adjacent topics to give me a full holistic understanding, it just sucks at it and is always short of helping me truly understand and intuit the subject matter.
qsera
Yes, this is my experience as well. At some point you would be better off find something written by a human, because AI would just take you in circles.
sivakon
This is an interesting usecase, and I want to learn more about your workflow. Do you also use Lean etc. for math proofs.
tim-tday
I’m the same way. But I took a bite and now I’m hooked.
I started using it for things I hate, ended up using it everywhere. I move 5x faster. I follow along most of the time. Twice a week I realize I’ve lost the thread. Once a month it sets me back a week or more.
lta
I repeatedly tried to use LLMs for code but god they suck. I've tried most tools and models and for me it's still way faster to write things by hand.
I'm a magical tool, it's almost like if I knew what I wanted to do ! Don't have to spend time explaining and correcting.
Also, a good part of the value of me writing code is that I know the code well and can fix things quickly. In addition, I've come to realize that while I'm coding, I'm mostly thinking about the project's code architecture and technical future. It's not something I'll ever want to delegate I think.
tehjoker
I use AI to discuss and possibly generate ideas and tests, but I make sure I understand everything and type it in except for trivial stuff. The main value of an engineer is understanding things. AI can help me understand things better and faster. If I just setup plans for AI and vibe, human capital is neglected and declines. I don't think there's much of a future if you don't know what you're doing, but there is always a future for people with deep understanding of problems and systems.
ehnto
I think you are right, deep understanding of systems and domains will not become obsolete. I forsee some types of developers moving into a more holistic systems design and implementation role if coding itself becomes quite routinely automated.
philipp-gayret
The atrophy of manually writing code is certainly real. I'd compare it to using a paper map and a compass to navigate, versus say Google Maps. I don't particularly care to lose the skill, even though being good and enjoying the programming part of making software was my main source of income for more than a decade. I just can't escape being significantly faster with a Claude Code.
> he can tell what it generates is messy for long-term maintenance, even if it does work and even though he's new to React.
When one can generate code in such a short amount of time, logically it is not hard to maintain. You could just re-generate it if you didn't like it. I don't believe this style of argument where it's easy to generate with AI but then you cannot maintain it after. It does not hold up logically, and I have yet to see such a codebase where AI was able to generate it, but now cannot maintain it. What I have seen this year is feature-complete language and framework rewrites done by AI with these new tools. For me the unmaintainable code claim is difficult to believe.
0x6d61646f
have you tried using AI generated code in a non hobby project? one that has to go to production?
it just allucinates packages, adds random functions that already exist, creates new random APIs.
How is that not unmantainable?
tokioyoyo
We use it daily in our org. What you’re talking about is not happening. That being said, we have fairly decent mono repo structure, bunch of guides/skills to ensure it doesn’t do it that often. Also the whole plan + implement phases.
If it was July 2025, I would have agreed with you. But not anymore.
vel0city
I used to experience those issues a lot. I haven't in a while. Between having good documentation in my projects, well-defined skills for normal things, simple to use testing tools, and giving it clear requirements things go pretty smoothly.
I'd say it still really depends on what you're doing. Are you working in a poorly documented language that few people use solving problems few people have solved? Are you adding yet another normal-ish kind of feature in a super common language and libraries? One will have a lot more pain than the other, especially if you're not supplying your own docs and testing tools.
There's also just a difference of what to include in the context. I had three different projects which were tightly coupled. AI agents had a hard time keeping things straight as APIs changed between them, constantly misnaming them and getting parameters wrong and what not. Combining them and having one agent work all three repos with a shared set of documentation made it no longer make mistakes when it needed to make changes across multiple projects.
philipp-gayret
Yes, all the time. Yes, those go to production. AI has improved significantly the past 2 years, I highly recommend you give it another try.
I don't see the behaviour you describe, maybe if your impression is that of online articles or you use a local llama model or ChatGPT from 2 years ago. Claude regularly finds and resolves duplicated code in fact. Let me give you a counter-example: For adding dependencies we run an internal whitelist for AI Agents; new dependencies go through this system, we had similar concerns. I have never seen any agent used in our organisation or at a client, in the half year or so that we run the service, hallucinate a dependency.
brysonreece
FWIW I mainly use Opus 4.6 on the $100/mo Max plan, and rarely run into these issues. They certainly occur with lower-tier models, with increased frequency the cheaper the model is - as for someone using it for a significant portion of their professional and personal work, I don’t really understand why this continues to be a widespread issue. Thoroughly vetting Plan Mode output also seems like an easy resolution to this issue, which most devs should be doing anyways IMO (e.g. `npm install random-auth-package`).
anonzzzies
We use it for 100s of projects and what you say hasn't happened for a while.
yojo
LLMs rarely if ever proactively identify cleanup refactors that reduce the complexity of a codebase. They do, however, still happily duplicate logic or large blocks of markup, defer imports rather than fixing dependency cycles, introduce new abstractions for minimal logic, and freely accumulate a plethora of little papercuts and speed bumps.
These same LLMs will then get lost in the intricacies of the maze they created on subsequent tasks, until they are unable to make forward progress without introducing regressions.
You can at this point ask the LLM to rewrite the rat’s nest, and it will likely produce new code that is slightly less horrible but introduces its own crop of new bugs.
All of this is avoidable, if you take the wheel and steer the thing a little. But all the evidence I’ve seen is that it’s not ready for full automation, unless your user base has a high tolerance for bugs.
I understand Anthropic builds Claude Code without looking at the code. And I encounter new bugs, some of them quite obvious and bad, every single day. A Claude process starts at 200MB of RAM and grows from there, for a CLI tool that is just a bundle of file tools glued to a wrapper around an API!
I think they have a rats nest over there, but they’re the only game in town so I have to live with this nonsense.
undefined
onlyrealcuzzo
I work at a FAANG.
Professionally, I have had almost no luck with it, outside of summarizing design docs or literally just finding something in the code that a simple search might not find: such is this team's code that does X?
I am yet to successfully prompt it and get a working commit.
Further, I will add that I also don't know any ICs personally who have successfully used it. Though, there's endless posts of people talking about how they're now 10x more productive, and everyone needs to do x y an z now. I just don't know any of these people.
Non-professionally, it's amazing how well it does on a small greenfield task, and I have seen that 10x improvement in velocity. But, at work, close to 0 so far.
Of the posts I've seen at work, they typically tend to be teams doing something new / greenfield-ish or a refactor. So I'm not surprised by their results.
tim-tday
This is wild. I’m on the other end.
I’ve probably prompted 10,000 lines of working code in the last two months. I started with terraform which I know backwards and forwards. Works perfectly 95% of the time and I know where it will go wrong so I watch for that. (Working both green field, in other existing repos and with other collaborators)
Moved on to a big data processing project, works great, needed a senior engineer to diagnose one small index problem which he identified in 30s. (But I’d bonked on for a week because in some cases I just don’t know what I don’t know)
Meanwhile a colleague wanted a sample of the data. Vibe coded that. (Extract from zip without decompressing) He wanted randomized. One shot. Done. Then he wanted randomized across 5 categories. Then he wanted 10x the sample size. Data request completed before the conversion was over. I would have worked on that for three hours before and bonked if I hit the limit of my technical knowledge.
Built a monitoring stack. Configured servers, used it to troubleshoot dozens of problems.
For stuff I can’t do, now I can do. For stuff I could do with difficulty now I can do with ease. For stuff I could do easily now I can do fast and easy.
Your vastly different experience is baffling and alien to me. (So thank you for opening my eyes)
ua709
I don’t find it baffling at all and both your experiences perfectly match mine.
Asking AI to solve a problem for you is hugely non-linear. Sometimes I win the AI lottery and its output is a reasonable representation of what I want. But mostly I loose the AI lottery and I get something that is hopeless. Now I have a conundrum.
Do I continue to futz with the prompt and hope if I wiggle the input then maybe I get a better output, or have I hit a limit and AI will never solve this problem? And because of the non linear nature I just never know. So these days I basically throw one dart. If it hits, great. If I miss I give up and do it the old fashioned way.
My work is in c++ primarily on what is basically fancy algorithms on graphs. If it matters.
enceladus06
What I've found Claude really helpful for is filling in the gaps. When you know vaguely how do to something like interpret data, but what other packages exist in xyz random technical domain? That is how I found for expample https://cran.r-project.org/web/packages/gggenes/vignettes/in... and Orthofinder when trying to teaching myself computational biology.
But sometimes even Claude gets stuck e.g. when I was trying to set up micropython via platformio running inside wsl2 on a windows 11 it got stuck setting up my ESP32 board.
wombat-man
Also at FAANG. I think I am using the tools more than my peers based on my conversations. The first few times I tried our AI tooling, it was extremely hit and miss. But right around December the tooling improved a lot, and is a lot more effective. I am able to make prototypes very quickly. They are seldom check-in ready, but I can validate assumptions and ideas. I also had a very positive experience where the LLM pointed out a key flaw in an API I had been designing, and I was able to adjust it before going further into the process.
Once the plan is set, using the agentic coder to create smaller CLs has been the best avenue for me. You don't want to generate code faster than you and your reviewers can comprehend it. It'll feel slow, but check ins actually move faster.
I will say it's not all magic and success. I have had the AI lead me down some dark corners, assuring me one design would work when actually it is a bit outdated or not quite the right fit for the system we are building for because of reasons. So, I wouldn't really say that it's a 10x multiplier or anything, but I'm definitely getting things done faster than I could on my own. Expertise on the part of the user is still crucial.
One classic issue I used to run into, is doing a small refactor and then having to manually fix a bunch of tests. It is so much simpler to ask the LLM to move X from A to B and fix any test failures. Then I circle back in a few minutes to review what was done and fix any issues.
The other thing is, it has visibility for the wider code base, including some of our infrastructure that we're dependent on. There have been a couple times in the past quarter where our build is busted by an external team, and I am able to ask the LLM given the timeframe and a description of the issue, the exact external failure that caused it. I don't really know how long it would have taken to resolve the issue otherwise, since the issues were missed by their testing. That said, I gotta wonder if those breakages were introduced by LLM use.
My job hasn't been this fun in a long, long time and I am a little uneasy about what these tools are going to mean for my personal job security, but I don't know how we can put the genie back into the bottle at this point.
goalieca
I can second this. I’ve never had a problem writing short scripts and glue code in stuff ive mastered. In places I actually need help, I’m finding it slows me down.
eranation
Wow, that's such a drastic different experience than mine. May I ask what toolset are you using? Are you limited to using your home grown "AcmeCode" or have full access to Claude Code / Cursor with the latest and greatest models, 1M context size, full repo access?
I see it generating between 50% to 90% accuracy in both small and large tasks, as in the PRs it generates range between being 50% usable code that a human can tweak, to 90% solution (with the occasional 100% wow, it actually did it, no comments, let's merge)
I also found it to be a skillset, some engineers seem to find it easier to articulate what they want and some have it easier to think while writing code.
ramraj07
I used to think that the people who keep saying (in March 2026) that AI does not generate good code are just not smart and ask stupid prompts.
I think I've amended that thought. They are not necessarily lacking in intelligence. I hypothesize that LLMs pick up on optimism and pessimism among other sentiments in the incoming prompt: someone prompting with no hope that the result will be useful end up with useless garbage output and vice versa.
oh_my_goodness
Exactly. You have to manifest at a high vibrational frequency.
bitwize
This is kinda like that thing about how psychic mediums supposedly can't medium if there's a skeptic in the room. Goes to show that AI really is a modern-day ouija board.
atmavatar
That sounds a lot more like confirmation bias than any real effect on the AI's output.
Gung-ho AI advocates overlook problems and seem to focus more on the potential they see for the future, giving everything a nice rose tint.
Pessimists will focus on the problems they encounter and likely not put in as much effort to get the results they want, so they likely see worse results than they might have otherwise achieved and worse than what the optimist saw.
cyanydeez
It's probably more to do with the intelligence required to know when a specific type of code will yield poor future coding integrations and large scale implementation.
It's pretty clear that people think greenfield projects can constantly be slopified and that AI will always be able to dig them another logical connection, so it doesn't matter which abstraction the AI chose this time; it can always be better.
This is akin to people who think we can just keep using oil to fuel technological growth because it'll some how improve the ability of technology to solve climate problems.
It's akin to the techno capitalist cult of "effective altruism" that assumes there's no way you could f'up the world that you can't fix with "good deeds"
There's a lot of hidden context in evaluating the output of LLMs, and if you're just looking at todays success, you'll come away with a much different view that if you're looking at next year's.
Optimism is only then, in this case, that you believe the AI will keep getting more powerful that it'll always clean up todays mess.
I call this techno magic, indistinguishable from religious 'optimism'
robbbbbbbbbbbb
Don’t know why you’re getting downvoted, this is a fascinating hypothesis and honestly super believable. It makes way more sense than the intuitive belief that there’s actually something under the human skin suit understanding any of this code.
wg0
This checks out logical speaking.
The FANG code basis are very large and date back years might not necessarily be using open source frameworks rather in house libraries and frameworks none of which are certainly available to Anthropic or OpenAI hence these models have zero visibility into them.
Therefore combined with the fact that these are not reasoning or thinking machines rather probabilistic (image/text) generators, they can't generate what they haven't seen.
threethirtytwo
No it doesn't check out. I think it's becoming abundantly clear LLMs learn in real time as they speak to you. There's a lot of denial and people claiming they don't learn that their knowledge is fixed on the training data and this is not even remotely true at all.
LLMs learn dynamically through their context window and this learning is at a rate much faster than humans and often with capabilities greater than humans and often much worse.
For a code base as complex and as closed source as google the problems an LLM faces is largely the same as a human. How much can he fit into the context window?
3form
You're observing this "paradox", because what you call learning here is not learning in the ML sense; it's deriving better conclusions from more data. It's true for many ML methods, but it doesn't mean any actual learning happens.
lta
It checks out if you take into account most developers are actually rather mediocre outside of places where they spend an insane amount of time and money to get good devs (including but not limited to FANG)
szatkus
That's why coding agents usually scans various files to figure out how to work in a particular codebase. I work with very large and old project, and Codex most of time manages to work with our frameworks.
therealdrag0
Huh? I have over a hundred services/repos checked out locally, ranging from 10+ years old to new. I have no problem leveraging AI to work in this large distributed codebase.
Even internal stuff is usable by the model because it’s a pattern matching machine and there should be documentation available, or it can just study the code like a human.
lta
Yeah that's still very far away from FAANG repos
waterTanuki
Not a FAANG engineer but also working at a pretty large company and I want to say you're spot on 1000%. It's insane how many "commenters" come out of the woodwork to tell you you're doing x or y wrong. They may not even frame it that way, but use a veneer of questions "what is your process like? Have you tried this product, etc." as a subtle way of completely dismissing your shared experience.
HeavyStorm
Same here. My take is that the codebase is too large and complex for it to find the right patterns.
It does work sometimes. The smaller the task, the better.
therealdrag0
Isn’t that fixed by having it create a plan, then you review it and say “x should do y instead”, it updates the plan, iterate then “build the plan”?
asdff
Can you elaborate on the shortcomings you find in professional setting that aren't coming up on personal projects? With it handling greenfield tasks are you perhaps referring to the usual sort of boilerplate code/file structure setup that is step 0 with using a lot of libraries?
humbleharbinger
I'm an engineer at Amazon - we use Kiro (our own harness) with Opus 4.6 underneath.
Most of my gripes are with the harness, CC is way better.
In terms of productivity I'm def 2-4X more productive at work, >10x more productive on my side business. I used to work overtime to deliver my features. Now I work 9-5 and am job hunting on the side while delivering relatively more features.
I think a lot of people are missing that AI is not just good for writing code. It's good for data analysis and all sorts of other tasks like debugging and deploying. I regularly use it to manage deployment loops (ex. make a code change and then deploy the changes to gamma and verify they work by making a sample request and verifying output from cloudwatch logs etc). I have built features in 2 weeks that would take me a month just because I'd have to learn some nitty technical details that I'd never use again in my life.
For data analysis I have an internal glue catalog, I can just tell it to query data and write a script that analyzes X for me.
AI and agents particularly have been a huge boon for me. I'm really scared about automation but also it doesn't make sense to me that SWE would be automated first before other careers since SWE itself is necessary to automate others. I think there are some fundamental limitations on LLMs (without understanding the details too much), but whatever level of intelligence we've currently unlocked is fundamentally going to change the world and is already changing how SWE looks.
jgilias
I saw somewhere that you guys had All Hands where juniors were prohibited from pushing AI-assisted code due to some reliability thing going on? Was that just a hoax?
erklik
https://www.aboutamazon.com/news/company-news/amazon-outage-...
About All Hands :
> Much of the coverage of the service incidents has focused on a weekly Amazon Stores operations meeting and a planned discussion of recent outages. Reviewing operational incidents is a routine part of these meetings, during which teams discuss root causes with the goal of continuing to improve reliability for customers.
This is something that's a part of every FAANG afaik. I know for a fact that there's no prohibition on pushing AI-assisted code. How would that even technically work? It'd basically mean banning Kiro/CC from the company.
> Only one of the incidents involved AI-assisted tooling, which related to an engineer following inaccurate advice that an AI tool inferred from an outdated internal wiki, and none involved AI-written code.
and this doesn't seem as "AI caused outage" as it was portrayed.
senderista
“outdated internal wiki” has to be responsible for so many AMZN outages…
etoxin
Not a hoax, saw it in the news. I'm not at Amazon but can confirm massive productivity gains. The issue is reviewing code. With output similar to a firehose of PR's we need to be more careful and mindful with PR's. Don't vibe code a massive PR and slap it on your coworkers and expect a review. The same PR etiquette exist today as it did years ago.
hn_throwaway_99
> I have built features in 2 weeks that would take me a month just because I'd have to learn some nitty technical details that I'd never use again in my life.
In the bucket of "really great things I love about AI", that would definitely be at the top. So often in my software engineering career I'd have to spend tons of time learning and understanding some new technology, some new language, some esoteric library, some cobbled-together build harness, etc., and I always found it pretty discouraging when I knew that I'd never have reason to use that tech outside the particular codebase I was working on at that time. And far from being rare, I found that working in a fairly large company that that was a pretty frequent occurrence. E.g. I'd look at a design doc or feature request and think to myself "oh, that's pretty easy and straightforward", only to go into the codebase and see the original developer/team decided on some extremely niche transaction handling library or whatever (or worse, homegrown with no tests...), and trying to figure out that esoteric tech turned into 85% of the actual work. AI doesn't reduce that to 0, but I've found it has been a huge boon to understanding new tech and especially for getting my dev environment and build set up well, much faster than I could do manually.
Of course, AI makes it a lot easier to generate exponentially more poorly architected slop, so not sure if in a year or two from now I'll just be ever more dependent on AI explaining to me the mountains of AI slop created in the first place.
skywhopper
It’s too bad, really. While it’s easy to get discouraged about such things, over the course of my career all that learning of “pointless” tech has made me a much better programmer, designer, architect, and troubleshooter. The only way you build intuition about systems is learning them deeply.
clintonb
> make a code change and then deploy the changes to gamma and verify they work by making a sample request and verifying output from cloudwatch logs etc
This has been a godsend over the past week while deploying a couple services. One is a bridge between Linear and our Coder.com installation so folks can assign the work to an agent. Claude Code can do most of the work while I sleep since it has access to kubectl, Linear MCP, and Coder MCP. I no longer have to manually build, deploy, test, repeat. It just does it all for me!
simonreiff
Mind my asking why job hunting and what you wish you could do in your day job that you're not?
3form
How do you deal with a risk of LLM generating malicious code and then running it? I suspect it's a bit more difficult to set it up tailorer to your needs in a big corp.
gerdesj
"I'm an engineer at Amazon"
Sanctioned comment?
jghn
> 10x more productive on my side business
Pretty sure the answer is here :)
gerdesj
Quite. On the face of it: possible career faux pas.
I own (with two other folk) my own little company and hire other people. I actively encourage my troops to have a bash but I suspect that a firm like AMZ would have differing views about what used to be called moonlighting. Mind you we only turnover a bit over £1M and that is loose change down the back of a sofa for AMZ ...
GeoSys
[dead]
wk_end
Around a year ago I started a new position at a very large tech company that I won't name, working on a pre-existing web project there. The code base isn't terrible - though not very good either, by-and-large - but it's absolutely massive, often over-engineered, pretty unorthodox, and definitely has some questionable design decisions; even after more than a year of working with it I still feel like a beginner much of the time.
This year I grudgingly bit the bullet and began using AI tools, and to my dismay they've been a pretty big boon for me, in this case. Not just for code generation - they're really good at probing the monolith and answering questions I have about how it works. Before I'd spend days pouring over code before starting work to figure out the right way to build something or where to break in, pinging people over in India or eastern Europe with questions and hoping they reply to me overnight. AI's totally replaced that, and it works shockingly well.
When I do fall back on it for code generation, it's mostly just to mitigate the tedium of writing boilerplate. The code it produces tends to be pretty poor - both in terms of style and robustness - and I'll usually need to take at least a couple of passes over it to get it up to snuff. I do find this faster than writing everything out by hand in the end, but not by a lot.
For my personal projects I don't find it adds much, but I do enjoy rubber ducking with ChatGPT.
abcde666777
Using these tools for understanding seems to be one of the best use cases - lots of pros, less dangerous cons (worst case scenario is a misleading understanding, but that can be minimized by directly double checking the claims being made).
In fact it looks like an arising theme is that whenever we use these tools it's valuable to maintain a human understanding of what's actually going on.
simonw
The majority of code I've written since November 2025 has been created using agents, as opposed to me typing code into a text editor. More than half of that has been done from my iPhone via Claude Code for web (bad name, great software.)
I'm enjoying myself so much. Projects I've been thinking about for years are now a couple of hours of hacking around. I'm readjusting my mental model of what's possible as a single developer. And I'm finally learning Go!
The biggest challenge right now is keeping up with the review workload. For low stakes projects (small single-purpose HTML+JS tools for example) I'm comfortable not reviewing the code, but if it's software I plan to have other people use I'm not willing to take that risk. I have a stack of neat prototypes and maybe-production-quality features that I can't ship yet because I've not done that review work.
I mainly work as an individual or with one other person - I'm not working as part of a larger team.
mudkipdev
Are you saying you're learning go because you've freed up time elsewhere or is AI helping?
Cyphase
How often do you find issues during review? What kinds of issues?
simonw
Usually it's specification mistakes - I spot cases I hadn't thought to cover, or the software not behaving as usefully as if I had made a different design decision.
Occasionally I'll catch things it didn't implement at all, or find things like missing permission checks.
snackerblues
>majority of code I've written
>has been created using agents
So you didn't write it
simonw
OK, the majority of code I've produced.
QuadrupleA
As a veteran freelance developer - aside from some occasional big wins, I'd say it's been net neutral or even net negative to my productivity. When I review AI-generated code carefully (and if I'm delivering it to clients I feel that's my responsibility) I always find unnecessary complexity, conceptual errors, performance issues, looming maintainability problems, etc. If I were to let it run free, these would just compound.
A couple "win" examples: add in-text links to every term in this paragraph that appears elsewhere on the page, plus corresponding anchors in the relevant page parts. Or, replace any static text on this page with any corresponding dynamic elements from this reference URL.
Lose examples: constant, but edit format glitches (not matching searched text; even the venerable Opus 4.6 constantly screws this up), unnecessary intermediate variables, ridiculously over-cautious exception-handling, failing to see opportunities to isolate repeated code into a function, or to utilize an existing function that exactly implements said N lines of code, etc.
slurpyb
It can only result in more work if you freelance because it you disclose that you used llm’s then you did it faster than usual and presumably less quality so you have to deliver more to retain the same income except now your paying all the providers for all the models because you start hitting usage limits and claude sucks on the weekends and your drive is full of ‘artifacts’, which incurs mental overhead that is exacerbated by your crippling adhd
And then all of a sudden you’re just arguing with the terminal all day - the specs are written by gpt, delivered in-the email written by gpt. Sometimes they dont even have the time to slice their prompt from the edges of the paste but the only thing i can think of is “i need to make the most of 0.5x off peak claude rates “
Fuck.
I got lots of pretty TUIs though so thats neat
vemv
Have you perceived a market shift for freelancers given the rise of AI coding?
It seems to me that sadly, paying for getting a few isolated tasks done is becoming a thing of the past.
QuadrupleA
No slowdown that I've seen - my style of freelancing is pretty long-term though, clients I've known and worked with for many years.
lazy_afternoons
Surprised to see HN being bearish on this.
I have 10 years of experience. I am a reasonable engineer. I can tell you that about half of the hype on twitter is real. It is a real blessing for small teams.
We have 100k DAU for a consumer crud app. We built and maintain everything in-house with 3 engineers. This would have taken atleast 10 engineers 3-4 years back.
We don't have a bug list. We are not "vibe coding" , 2 of us understand almost all of the codebase. We have processes to make sure the core integrity of codebase doesn't go for a toss.
None has touched the editor in months.
Even the product folks can raise a PR for small config changes from slack.
Velocity is through the roof and code quality is as good if not better than when we write by hand.
We refactor almost A LOT more than before because we can afford to.
I love it.
karmasimida
HN is in denial, which is understandable
AI is already better at understanding code than 99.99% of human, the more I use it the more I believe this is true. It can draw connections between dots far quicker and accurate than a human could ever be.
At very least, AI is going to be a must even as a co-supervisor to your project
What in doubt right now, is whether AI can manage a codebase fully autonomously without bring it down, which I doubt it can at the moment. Be it 4.6 or 5.4, they always, almost always, add code instead of removing them, the sheer complexity will explode at certain point.
But that is my assessment for models TODAY, who knows where they will end up being in 6 months. AI is entering the recursive self improvement phase, that roadmap is laying in front our eyes, what it can and would unlock is truly, truly unpredictable.
I am both intrigued and scared.
iExploder
> AI is already better at understanding code than 99.99% of human
not to nitpick here, but AI does not understand code. AI (LLMs) are token predictors pr at best sophisticated pattern matching in huge search space...
qwertybrah
[dead]
civvv
The RAG models are very competent at programming. I am worried about my job as a SWE in the near future, but didn't the MIT paper about a week ago pretty much confirm that width-scaling the model is about to (or has already) stopped giving any measurable increase in quality because the training data no longer overfills the model?
Any authentic training data from pre-LLM's is assumed to have been used in training already and synthetic or generated data gives worse performing models, so the path of increasing its training data seems to be a dead end as well?
What is the next vector of training? Maybe data curation? Remove the low quality entries and accept a smaller, but more accurate data set?
I think the AI companies are starting to sweat a little, considering the promises they have made, their inability to deliver and turn a profit at its current state and the slowing improvements.
Interesting times! We are either all out of jobs or a massive market crash is imminent, awesome...
cauefcr
Different architectures, different RL training loops, maybe memory modules [1][2] as part of the architecture, focusing on efficiency, the giant troves of data we're generating by using claude code/gemini-cli/opencode, there's lots of research to be made.
[1] https://research.google/blog/titans-miras-helping-ai-have-lo... [2] https://github.com/deepseek-ai/Engram
j3k3
Link? Genuinely curious to check it out.
bdangubic
100k DAU - you’ll lose 98% within 6-9 months once 1-2 person team clones it as sells it for 10% of what you are charging
vdfs
That was always a possiblty even before AI, what's hard to clone is how they got those users
bdangubic
possibility yes - reality often no due to cost that would have to be incurred to make this happen. the "how they got those users" is the easy part if you offering is same(ish) at a fraction of the cost.
make_it_sure
most developers are still in denial. Many are afraid of job loss or the corporations are forcing AI without clear scopes and proper implementation, which results in a mess. Small teams for small-medium products are productive as hell with AI.
greenpizza13
I work at a very prominent AI company. We have access to every tool under the sun. There are various levels of success for all levels — managers, PMs, engineers.
We have cursor with essentially unlimited Opus 4.6 and it’s fundamentally changed my workflow as a senior engineer. I find I spend much more time designing and testing my software and development time is almost entirely prompting and reviewing AI changes.
I’m afraid my coding skills are atrophying, in fact I know the are, but I’m not sure if the coding was the part of my job I truly enjoyed. I enjoy thinking higher-level: architecture, connecting components, focusing on the user experience. But I think using these AI tools is a form of golden handcuffs. If I go work at a startup without the money I pay for these models, I think for the first time in my career I would be less likely to be able to successfully code a feature than I could last year.
So professionally there are pros and cons. My design and architecture skills have greatly improved as I am spending more time doing this.
Personally it’s so much fun. I’ve made several side projects I would have never done otherwise. Working with Claude code on greenfield projects is a blast.
abcde666777
I think people get a bit paranoid about coding skills atrophying. I had a period where I stopped programming for multiple years and it really only took a month to get back into the swing of things when I returned, and most of that was just re-jogging my memory on the syntax and standard library classes (C++ at the time).
georgemcbay
A month is quite a long time compared to "I can just do this at-will from neutral at any time".
...particularly in situations where you might have to navigate a change in jobs and get back to the point where you can reasonably prove that you can program at a professional level (will be interesting to see how/if the interviewing process changes over time due to LLMs).
cloverich
i also worry but am also shocked how far a single $20 sub gets me on side project. i pay for 3 (cc, codex, gemini) but am almost never going beyond cc, even when im merging several prs a day.
kreyenborgi
Net negative. I do find it genuinely useful for code review, and "better search engine" or snippets, and sometimes for rubber ducking, but for agent mode and actual longer coding tasks I always end up rewriting the code it makes. Whatever it produces always looks like one of those students who constantly slightly misunderstands and only cares about minor test objectives, never seeing the big picture. And I waste so much time on the hope that this time it will make me more productive if only I can nudge it in the right direction, maybe I'm not holding it right, using the right tools/processes/skills etc. It feels like javascript frameworks all over again.
christophilus
Same. I vacillate between thinking our profession will soon be over to thinking we’re perfectly safe. Sometimes, it’s brilliant. It is very good at exploring and explaining a codebase, finding bugs, and doing surgical fixes. It’s sometimes good at programming larger tasks, but only if you really don’t care about code quality.
The one thing I’m not sure about is: does code quality and consistency actually matter? If your architecture is sufficiently modular, you can quickly and inexpensively regenerate any modules whose low quality proves to be problematic.
So, maybe we really are fucked. I don’t know.
Get the top HN stories in your inbox every day.
Comment sections on AI threads tend to split into "we're all cooked" and "AI is useless." I'd like to cut through the noise and learn what's actually working and what isn't, from concrete experience.
If you've recently used AI tools for professional coding work, tell us about it.
What tools did you use? What worked well and why? What challenges did you hit, and how (if at all) did you solve them?
Please share enough context (stack, project type, team size, experience level) for others to learn from your experience.
The goal is to build a grounded picture of where AI-assisted development actually stands in March 2026, without the hot air.