fhd2

But would you argue that it is neither about short term nor long term financial gain? Cause that was my point. It really always is about profit one way or another, in my experience, with barely any exception I can think of.

Well, it's either about "profit now" or "profit later". The latter one can feel a lot like it's not about profit.

Yup, I rather meant to point out that even if all you want to do is sell access to the model on your hardware, you can _still_ do so up until a non-trivial scale.

Unless the total (!) revenue of the provider is under 20 MM USD, eh? Not exactly fully open, but not _that_ harsh.

I'm in Germany, and I usually just find a good option outside Amazon, then look for the best deal, which is rarely Amazon. Thanks to express checkout (Google/Apple Pay or PayPal is almost always an option), it's pretty much zero hassle.

BUT, if I care more about getting something real fast than the quality, or if I don't want to pay 5 EUR for shipping (especially cheap stuff), I'm not aware of any competition Amazon has here. I use it quite a bit still, for these two use cases.

Thought it was funny, might be wrong about that :P

Well, that's today. Also without actions and pull requests.

That's my experience as well. Some folks like to dabble, but for the most part they have other stuff they want to focus on, they don't want to spend the time to ensure it's good and it works.

I'm not saying an agent couldn't make sure it's good and it works, but there are a lot of aspects to that: It's compliant, it's secure, the right data is stored, the right compromises are made for sometimes mutually exclusive business priorities, UX is decent, it's responsive, it scales, hosting is economical, it has the right tests, monitoring for the right risks, and so on and so forth. A lot of devil in the details, and the context required to know what to focus on is rather vast.

If anything, I see this work getting harder, since the discipline around minimal scope is getting weaker, leading to a complexity explosion affecting all of these aspects.

It usually comes down to how successful a business is. Just starting out with little to no funding? They'll want to dabble themselves. But if there's money, it can and will be exchanged for time. Easier prototypes and MVPs means faster business model validation, which means higher willingness to invest. I do see this happening.

Will we always need to know programming languages for doing this? Maybe not, but it doesn't matter. If you make a machine do stuff for a living, you're a programmer. How exactly you do it doesn't affect that in my mind.

Mandatory feature.

That sounds pretty frustrating. I wonder if it wouldn't be a good approach to, say:

1. Make the exam the only thing that determines the grade (or technically, proves that the student learned enough);

2. Give out test assignments students can do, but don't have to;

3. Offer students for the professor to grade their work if they want to get some feedback on how they're doing.

Would that lead to less time wasted by tutors grading AI generated answers? From where I stand, students should be allowed to prepare for exams any way they like, with or without their professor's help. If they managed to learn, they pass.

If you blacklist them just for yourself, the sites don't care, no economic impact. If Google does the same, they'll work around. There's lots of spam-only sites, but they also often cleverly avoid detecting by mixing in content ranked as valuable/legit. SEO also certainly "attacked" otherwise legitimate sites like Reddit, although what GEO is doing there is far worse I imagine.

I interact with plenty of companies - not a statistically relevant amount, but dozens. I'd put this at about 50% right now. I certainly deal with a not insignificant number of meat proxy CEOs, which is highly annoying for the people worrying about the details. But it's not impossible to work with. I just accept the overhead and hope it gets better.

Hang in there. I've yet so see an even modestly successful business that didn't have this or some other bleak problem at one point in time. I'm sure they exist, I just haven't seen one yet :D

My gut says this one can be turned around. Even if 90% of people just happily prefer ChatGPT for this (which I'd doubt), the remaining 10% is still a lot to fund a small venture very well. Just gotta find them effectively.

Even the "giving up" route can be pretty juicy, there's companies that pay decent money for a "distressed asset". And there's also investment firms (like mine) that go for a minority stake and help turn things around.

Just speculating (I use zero automation when recruiting), but automated candidate ranking systems could well still look for signals that were valuable years ago, but no longer are. It's not a given that people using this stuff even understand the criteria under the hood.

Same story as SEO, I presume. Google surely understood their ranking criteria deeply, yet couldn't exactly win that fight.

Nice analysis! I would perhaps add that LLMs are pretty helpful at (2) in my experience, as long as you don't turn your brain off. (As a Google/SO replacement, mainly.)

I have about three modes, working with LLMs:

A) Just use them for (2), do (3) and (4) myself.

B) Just use them for (2) and (3), but with very fine grained instructions - to the point where it barely saves me time.

C) Use them for (2), (3) and (4), but with this workflow, it quickly degrades into slop for me.

I still get the best results with (A), but for a quick PoC, (C) is hard to beat.

What's hilarious (or sad) is that I see a lot of people throw them at (1) as well, which seems to have a high chance of inducing AI psychosis and creating software nobody understands, let alone needs.

With GenAI, we can now produce something without caring about it - or while caring about it very little.

And the parts we don't care about aren't necessarily worse, they are just... arbitrary. Could be good, could be bad, no one knows, because no one really cares.

I find that if I care about something a lot, it's a pretty similar time investment than pre LLMs. And it makes me feel invested and proud in the result, motivated to show it and improve it.

If I care about something very little, in the past I just wouldn't have done it at all. Now I might, but I feel that same disconnect you mentioned.

I think being strategic in what we do and do not care about is likely the key skill we'll have to build to actually make the best of the tech.

Not entirely my experience. LLMs can be good at two things, but unable to be good at them _at the same time_. I've had some success with getting decent simplification suggestions out of them.

Still ironic, of course.

I never went into an Apple store, but the phone support issues I had (even a pretty complex bug on their end that locked me out of my account) where handled swiftly and well, and I spoke to humans that seemed to know what they're doing. I could say this should be the default and I shouldn't be impressed by it, but unfortunately it's top 10% based on my experience. I'm in Europe though, maybe they do things differently here.

Apple has pretty stellar support in my experience, and while I don't buy Apple stuff for other reasons, this is a strong attractor. For B2B products/services, it can be a huge differeniator as well, in my experience. But it was really just an example, it might not be the call most companies make. Maybe most will still just invest in being faster or cutting costs. I just hope that's not how it's gonna be.

Right now, the majority of companies I interact with seems to look at their now (let's assume) increased capabilities, and turns up the speed. But there's other knobs to turn up instead. Robustness, quality, user experience.

Optimistically, I think we might just enter a new phase at some point, where businesses make very conscious decisions about what to invest in to differentiate. Like, everyone has support bots now. Where do we invest, absolutely stunning human support? Or is support not that key and we should invest into being the most reliable? The most enjoyable to use?

One can hope.

Good example. Transitioning from an outdated framework to a modern (or sometimes "slightly less outdated") one is probably one of the few situations where you do not want to change semantics at all.

And in my experience, these are _dangerous_. People go into "while we're at it..." mode, and it quickly turns into a big 2.0 kind of thing that takes forever.

I would argue that LLMs can speed this kind of thing up, but not by an order of magnitude or anything, just a bit. Unless there's high risk appetite.

Depends on how much you're asked to burn I guess. Until one point, it's actually helpful. Then there's a point where you can do stuff that's semi helpful but doesn't get in the way. But then I would imagine you reach a point where you have to come up with a token burn strategy and some kind of narrative for your manager that's in line with it. I bet at that last point, it gets taxing.

Probably like eating. Having to eat less to loose weight isn't great. Eating anything you want without worries is great. Having to eat more than you want to gain weight, not great.

AGENTS.md is fun too, solid AI bro stuff:

> Code like Anthony Fu and Evan You

> default to the highest quality modern code the legends would ship

What a time to be alive.

If it's profitable, they would. Companies aren't moral, with very rare exceptions. Legal risks and costs are just part of the overall equation.

But that's not in defense of what Karp said, or that slop tweet summarising it. That's... wild. I regret reading it.

I think it's primarily a filter. Candidate filters are pretty much always silly, even common ones based on degree/grades etc. But with a lot of candidates on the market, using some filters reduces the list to a manageable amount.

Personally, as someone with a German company and a good chunk of German clients, I'd argue it _does_ help a little. Occasionally. But by and large I'm far more interested in the candidate's English proficiency.

I'd argue it's pretty much like monitoring, which certainly benefits from multiple people seeing the same stats and alerts. I agree it's at odds with CI/CD and should probably not block anything, like deterministic checks commonly do.

One fascinating thing about LLMs is the degree of evangelism it inspires in some. You can explain some of that with paid micro influencers, people invested in the success of AI, consultants looking for workshop opportunities and all that, but I know enough people with no skin in the game at all, that turned into very vocal advocates.

I think to some degree, that effect is also at play here. CEOs, product managers etc are simply amazed, and want to spread the good news. I doubt they can even _comprehend_ that others might not be as excited as them.

SEO is definitely a relevant component still, AI mode does a traditional search under the hood. Gaming the system to rank high on the index is still going to matter. But it'll be less about tricking people into clicking and more about tricking LLMs into considering the information relevant and authoritative. For someone using traditional search, I'd wager that would actually improve the results a bit over time.

Then again, SEO gaming got a whole lot cheaper with LLMs, spammers spam even if there's not a great return, as long as it's cheap for them.

Yeah, plus it's a bit... single minded. A static single page site is _quite_ "agent ready". Scores 0 here. It's not like it'll need an MCP or whatever.

So AI generated code doesn't benefit from stable foundations maintained by third parties? Fascinating take I don't currently agree with. Whether it's AI or hand written, using solid pre-existing components and having as little custom code as possible is my personal approach to keep things maintainable.