Unfortunately, this is the type of statements we can't verify. I'm not sure why these types of news are still coming out when we all have AI at work.
Whenever someone does such a huge drastic change like this, it's by ignoring a large chunk of code that most people were afraid to touch for good reasons. Now, that code is gone, AI is celebrated, things will break, people will work very hard in the background to fix it, with no fanfare.
> I'm not sure why these types of news are still coming out when we all have AI at work.
Because OpenAI is burning $15 billion/year, and outrageous stories like those get parroted in the media. It's free marketing for a company desperate to get middle management to believe that a $500/mo subscription is absolutely crucial for every single employee.
ugh just recently I saw a news article gushing about how AI had helped with some health/medicine study, and various patient organizations were all like "oh yeah this is a Good use of AI" and then I click the link to read the study and it's decision trees and clustering on a tiny dataset that you could analyze with a ten year old laptop.
I mean, sure at some point decision trees and clustering were called "AI", but the way the article was written you'd think OpenAI and Anthropic were responsible for the advancement of medicine.
Decision tree learning is a form of machine learning and machine learning is subject of AI research. Same for clustering. Why are you saying they "were" called AI? They still are. Maybe everyone now synechdochically calls LLMs AI but that doesn't mean other kinds of AI are not AI.
AI research is not over yet. If OpenAI and Anthropic fail to bring on the Singularity, what are we going to call the continuing research on AI? Are we going to call it something else than "AI" because that was taken by LLMs? Does that make any sense at all?
I kind of agree, though I find "AI" to be an almost useless term, in particular these days. I dislike the current state of discourse where AI is used to mean inhumanly large models, so I tend to use a specific term like LLMs instead of AI. But the point here wasn't whether this or that form of ML is in fact AI. The term "AI" was almost never used in mainstream media until a few years ago, when it suddenly it was all over the place
https://trends.google.com/trends/explore?date=all&q=AI,Biden...
and nearly without exception referring to very large generative text/image models. So when a news article in 2026 talks about AI without specifying that it's nothing to do with ChatGPT etc. (and in fact using a very different method requiring no large pre-trained model and no dependence on large companies), I find it highly misleading.
yeah but you'd need skills to operate the 10 year old laptop. probably have to copy and paste and understand excel and maybe whatever the hell a confidence interval is
this allows said middle managers to do the analysis themselves and feel mostly confident in the results.
I can for sure say that this kind of statement is exaggeration if not a blatant lie. It may work for PR but it gives a wrong impression/ideas. Now engineers should be ready to be questioned by their management for all the rewrites that they have been estimating in months if not years.
> But at the rate we were going, we were still roughly five years from finishing.
Everyone has seen how this sort of thing comes about: it's meaningful work for engineering but never business/product critical so it just drags along.
Seems like this time around someone just went "I wonder if we could do it this way" and it worked. Perfect example of ditching sunk-cost and starting from scratch. Great outcome for them.
This doesn't "verify" anything. I can't see the code before, I can't see the code after. I can't verify they do the same thing. It won't be updated if there's a long churn of issues and breakage coming from this work plaguing the team for years. The only thing it verifies is that Asana did indeed make the claim, which I don't think anyone was doubting.
> Back in 2022, we set out to migrate Asana's frontend test suite off Enzyme, our aging testing library, and onto React Testing Library (RTL).
Your telling me they had a full team of engineers at Asana, doing nothing but rewriting tests for 4 years until AI came along and did the last year in a couple days? I'm extremely doubtful. I don't doubt for a minute they were one-track to take 5 years, but not because it was 5 years of engineering effort for humans.
Far more likely they finally cleared up some tech debt they had been plugging at off and on for 4 years, then a PR flack got ahold of it and it became a breathless "AI did 5 years of work in a couple days".
I'm not sure I agree with the extremely easy to verify part. How do you validate that your new test suite covers the exact same edge cases as your old?
But yes, LLMs are great at tests. I'm not doubting that an LLM migrated some legacy tests, or that it was a task taking a long time. I am doubting the way it was presented in the article, that this was taking a full team of engineers dedicated to nothing but this, 5 years to accomplish (and presumably already spent 4 million working on this, since the total estimate was 6 million, and they've been at it for 4 year).
I think some PR flack got ahold of the fact that an LLM wrapped up migrating some legacy tests that a team had been slowly chipping away at for 4 years, between their normal feature work, and were on track to finish in 5, then wrote it up like it was that teams entire focus instead of a piece of tech debt.
That makes far more sense to me than spending millions for a team of engineers dedicated to nothing but rewriting an existing test suite.
Unit testing can get you some of the way, but its not a full-on all-case guarantee. Sometimes the code is encapsulating some particularly complex system/behaviour. Sometimes the reason is interop/compatilibity issues or some kind of politics.
P.S. you managed to introduce a typo in your quote (were –> where)
Or the reason is that something is just unfeasible with the existing architecture, or the reason is that building this would break important technical assumptions... All kinds of things.
Exactly. You can simply close all open tickets with <WONTFIX> and claim that you've cleared 5 years of engineering work in 5 minutes. It doesn't mean anything.
It’s just sales claims.
Do you see that “contact sales” button?
You’re not the target of that.
Deranged AI-psychotic C-suites with a two-digit IQ and too much confidence are the target.
Phrased this way, my aspiring product leader hat activates: that 5 years of work cannot be of much value to a company worth billions because a 5 year estimate is the lowest possible qualification you can make. Its a “will never happen” shelving.
all work is stack ranked against other opportunities. no matter how many teams help to parallelize, leadership basically comes down to ranking the stack correctly. Valuable work will be at the top by any means necessary. “mvp”, probes, task forces, code yellows, and so on.
edit: it’s even a backhanded compliment to agentic coding. Asana a public company considers a $12k outlay for something they’d never spend real resources on worth a case study =|
Apparently the task was to remove dead weight. Haven't read what they replaced "Enzyme" with.
I did subagent based removal tasks a few times. These were the ones that required the least amount of input or thinking from me, because the requirements are abundantly clear. "Remove this part of the code without breaking any other part or by porting the tests done with it onto another system."
In these situations the code acts as the bookkeeping ledger itself, and coordination complexity is almost a no-brainer.
either it was a hail mary 5 year change-the-market offering, or it was something that staff can "get around to" whenever they have the cycles, and would, at current rates, take like 3-5 years to do.
was the tool worth 12k? if it's not a line in capex or opex budgets it's value is $0
One thing from experience having worked on a project 4 years ago, 2 of us spent a a couple of months investigating and coming up with a solution.
I tried it in AI and virtually did it all in a few hours! Magic?
Well I had gone through it to explore the problem space, and we had found a working solution which just petered out. But I still remembered the learnt lessons, and I was just eager to see what it would be like to use DuckDB on it.
My point being these second passes on projects often skip the hard won knowledge part.
If you ask Sol or Claude how much time it will take to implement a plan they just came up with, they usually advise a timeframe in the weeks or months - assuming, I suppose, that human programmers will be building it. And then you ask the model to just "do it" and it takes an hour or two. I always find this entertaining.
It does, assuming "world" is made up of humans (which it is, in its training data). It does however not have any model whatsoever of itself, apart from some superficial "I am a helpful assistant built by $X".
I find that the models have increasingly started throwing out made up amounts of time around, as if they wouldn't be aware that the user is already using Claude Code or Codex.
It's like every time they make a plan, there's something about things taking "a week or two", "month of focused work", or whatever.
This is something that would've been RL'd out a long time ago if it wasn't great for business.
A model isn't "aware" of anything. It's trained on things, and that training doesn't include time estimates for working with AI because that didn't exist yet.
What I was going after with "aware", is that the actual people working at the companies, training the models, are aware that people aren't mostly going to be implementing the plan by hand, if they've already made the plan in Claude Code or Codex. As for a specific Claude / GPT instance, "aware" would definitely be the wrong word choice there, but the instance does have its stats and environment information in its context window, unless you specifically remove it.
Either way: Training does include estimates on working with the model, and adjustments of the model itself based on that. That's literally what RLHF is.
It's straightforward to have a portion in post training that aims specifically at the model being able to give better estimates on how long that model takes to complete a certain type of task.
In all honesty I don't think it's possible to put this into the model because it's not actually sufficiently understood yet by anyone or anything. AI capabilities are changing too fast, and so is the way AI is used (worst case, even including financial limitations that are now increasingly appearing.)
Every part of training a model have the timing data observed + saved on multiple different axis, it's one of the most inherent parts of the training process.
When the post training run is nearing its cutoff point, there's a massive amount of data on how long coding tasks take to complete by that model in the golden format of "task -> time task took to complete", separable to whatever amount of subtasks, in the same format. With the parts from the end of the dataset being useful for evaluating the finished model's capabilites, whether that data is then fed back into another step in post training or not.
Completely separate from even the actual training: If you have a model proactively giving estimates that are an order of magnitude wrong, you can already fix the worst of it as of this moment by just changing the system prompt. It's a dirty fix, but it's the type of fix that has been used by Anthropic and OpenAI since forever when a model is dishing out blatantly wrong outputs.
The misunderstanding might be that I'm thinking about how long it takes a human + an AI to get something done, not an AI by itself. My current understanding is that it's still not generally viable to have LLMs perform architecture/design tasks (it might work in the short run but degrades roughly exponentially in the long run, accruing serious tech debt). Most things I'd consider actual "projects" involve at least some level of design work as well, and that'd need to be included in the estimate.
(And also prompts will need to be refined, etc., unless you have amazing prompt skills it won't immediately deliver what you wanted. Even if it did, the work to put together the AI inputs also needs to be included.)
⇒ I don't think it's easily possible right now to give a time estimate for **the full picture of** a project to be implemented with AI assistance.
> For comparison: the previous plan was expected to take at least five years and estimated to cost roughly $6M.
If that's an estimate from an LLM, those have always been way off for me. I'm constantly amazed whenever something that an agent estimates would take weeks ends up being completed in an hour.
The LLM is reasoning about estimates from its training data... which is to say, from human engineering timescales.
I suspect the labs could improve the models such that they are estimating these sorts of things but they don't prioritize doing so (or perhaps RLHF selects it away) because, as you say, it feels amazing to do a week's worth of work in an hour.
I can easily clear lots of low-priority engineering work quickly using AI. It takes a bit longer if I want to do a good job.
But here’s the rub: this is, basically by definition, low-priority engineering work. Those fixed papercuts are nice to have fixed but they do not necessarily add a lot of value.
And it’s all too easy to lose value by doing this. For example, current LLMs really really like adding test cases, and a lot of those test cases have basically no value, and carrying them around is not free.
It has it risks too. I've used chatGPT for a feature I am working on. It works, I have already resolved some bugs the it didn't pick up. However, the code quality is odd, I've had some hardcoded types as it is unlikely they will ever change. For some of them chatGPT created enums but only for some.
I also realized that review generated code sucks, I can generate a whole app quickly, but I have no understanding of small decisions. It is extremely difficult to wrap your hand around hundreds of lines of code written by someone else.
If you mean chatgpt.com, try with a real “agent” harness. The exact same model on chatgpt.com frequently performs much worse than it does with a proper harness, not to mention that the chatgpt.com frontend itself is quite bad.
codex-rs, popularly known as Codex or maybe the Codex CLI, is a fine choice.
Is this the react test framework enzyme? To me, $6 million and 5 years seems like a very high estimate for this even without any automation (AI or otherwise) for even relatively large applications. How could they possibly have a test suite this large?
The British government cleared a large chunk of 5 years of the NHS (hospital) waiting list by, in part, just removing people from it if they'd been waiting too long.
People may think it’s an overstatement. But I think it’s not.
This kind of projects are where AI is most helpful. Long tedious and highly testable projects like ports or legacy system replacements where humans have to grind through millions of lines of code without really thinking are the perfect target for AI.
I once did a C to C# port of several math libraries, and while I was able to automate most of it, it still required a ton of manual work. I bet if I had Claude, I could do what I did in 3 months in a day or two.
This is what I was thinking too. Ports are probably where AI has the one of the biggest time savings over humans.
Although also agree, that even if human's did it, they probably would try to automate as much as possible and speed up development, and maybe that original 5 year estimate was if it was done completely manually.
There are countless years man work awaiting in the issues of abandoned projects. Nobody is getting fired because of them. But new people won't be hired if managers believe in marketing stunts like this.
I would love to know the responsible people at Asana who approved this joint PR. Like, do you not get the blatant second degree negative implications of what’s stated here, albeit manipulated and hyperbolic, for a project management software?
The five year project likely represents a bunch of engineers that are stuck in meetings and coordination for 75% of their time, then you staff the project with more people (because you need four times as many) then they all need more meetings to coordinate too. And then they need to write a lot more documentation in the process because in that five years the people working on the project will all turn over at least once if not 2-3 times. Never mind critical people getting pulled off the project to work on other more important things on a monthly basis as is common at large firms.
It's entirely possible for a 2-3 month project if one senior person was left the hell alone to do it to take a dozen people five years.
> An engineer checked progress twice a day and reviewed every proposed change.
It doesn’t sound like that much changed. If one engineer can really review all the changes in under two weeks, the original estimate of 5 years of engineering work has to be waaaaaaay off.
When everything is “estimates”, anything sounds amazing.
The impressive thing here is that Asana has such a poorly run engineering org that replacing their testing framework was estimated to take 5 years and $6 million.
The fun part is how the article just says they removed "Enzyme" and there is nothing else mentioned that can make you gauge the technical complexity of what was done here
It does when you have a team of 100's of engineers every change has to pass through, 3 layers of management, SRE teams, DevOps YAML engineers, product management and a partridge and a VP.
1. I am not sure what your profession is, but I spend most of my reading time at work, with code, emails and so on. Unfortunately I don't have much time to read for recreation these days.
Comparisons such as 5 years and $6M are meaningless. Instead, they should compare with other AI, such as how much it would have cost using self-hosted open-weight models, or using a API that offers open-weight models etc.
The latter sounds about as attractive to me as the former, actually more because my customers are mostly hoping for their bugs to be fixed , not for the latest novel feature.
These conclusion can be obtained by any individual adult with a command of the English language and average written text comprehension abilities within 2 minutes. That’s how long it takes to read and understand the case study.
So, why did you surrender your cognition to an LLM? Do you have some form of mental impairment?
Whenever someone does such a huge drastic change like this, it's by ignoring a large chunk of code that most people were afraid to touch for good reasons. Now, that code is gone, AI is celebrated, things will break, people will work very hard in the background to fix it, with no fanfare.
reply