Shopify CEO: "AI usage is now a baseline expectation"
twitter.com
twitter.com
In the past, it was hiring (or at least the appearance of hiring). But that's an expensive signal and not sustainable for long in a post-ZIRP period. On the other hand, bullshitting about AI (and before that, blockchain) is much cheaper in comparison, and seems to do just as good without actually needing to hire anyone or even pretend to.
Whether AI is actually used, or helps the bottom-line, is irrelevant (it's not possible to conclusively say whether a piece of code was authored by AI or not, so the subsequent narrative will be tweaked as necessary to fit the market conditions at the time).
I have to wonder when some startup will attract talent with a tag line like "We use AI like we use text editors: Only where it makes sense."
Their actual AI is trash, it just there to pad their resume for investors.
I tried few days ago their AI assistant in help center and admin and it hallucinated answer to my question (pointed me to non-existing setting), when I told it it doesn't exist it just returned a bit reworded answer (still pointing me to non-existing settings section).
Not kidding, but I'm actually afraid people will check AI usage and start nagging us that:
> "You are slow because you don't use enough AI. Look at Tom (single 28 yo guy working every weekend on Adderall), he uses AI and he is fast. Gonna add a note to check up on your AI usage metrics in a month, hopefully it will improve".
Our company has Cursor, which I sometimes use, but 1. for lots of tasks, the most precise language is the programming language, 2. I don't love it, prefer other editors, and I go for in-browser search + AI.
If this letter was published by my CEO, I would either 1. ignore it, as CEOs are often out of touch when it comes to actual day to day work or they need to jump on the AI train to show how visionary they are even if they are aware of the limitations, 2. start looking for a job, because honestly, today it's a letter like this, in 3 months, you get a negative performance review because you didn't send enough queries to Cursor.
This is my problem with AI, or "vibe coding" or whatever you want to call it.
We already have many language(s) for expressing what we want the computer to do, and I find that natural language is the most difficult one to accomplish that goal.
Sure, LLMs might be able to whip up a fancy landing page and some basic API really quick, but when it comes to solving harder problems I find it incredibly frustrating and difficult to try and prompt in English, when the programming language I'm working in is already well suited to that task and is much more explicit in what I want.
Maybe I'm using it wrong, but it's way less cognitive overhead for me to just type for for x,y := range something {} than it is to try and prompt "Iterate over this list of values...etc."
When I'm programming, I'm not thinking in English
I'd say that prompting "Iterate over this list of values...etc." is definitely using it very wrong (autocomplete should more or less handle that sort of thing anyway). Prompts should be more in line with "write a C++ function that can parse XML files that look like this (upload a few sample files) and return the data in this struct (copy and paste struct from your header file)" followed by "write a set of unit tests for this function". You then check that the unit tests look reasonable, add any other things you feel you should test for, make sure the generated code passes the unit tests, and then check it in.
The "look at Tom" thing isn't new, there have always been sloppy engineers that crank out bad code faster than their peers and got praised for it while slowing everyone else down because they had to actual debug and integrate the garbage.
I'm also not a fan of how productivity expectations seem to getting worse because the people in the business side read that this makes programmers 150% more productive. They probably do write more code but if the shelf life of that code is worse and there's less knowledge in the organization about that code because the engineer was leaning on a stochastic tool, how much more are you gonna spend maintaining and rewriting that software? Just seems like we're all super excited to make crappier software faster.
is that better or worse than ketamine for weekend productivity
Oh, oh and I almost forgot, I'm also going to need you to go ahead and come in on Monday/Friday too. Okay. We lost some people this week so we need to play catch up. Thanks!
Or because you spend most of your time unravelling the crap code that AI produces. Creating apps with AI is one thing. Maintaining them is another.
Am I fundamentally missing something about the experience upper management has with AI versus the experience of a mid/senior level developer in the trenches? I also have access to Cursor and a few other LLMs.
They're handy for bouncing ideas around and for some simple tasks, but I've hardly felt like it was a force multiplier.
When it comes to coding, the amount of times it's hallucinated a function that didn't exist or used a deprecated implementation has felt like a net neutral at best.
> We will add AI usage questions to our performance and peer review questionnaire.
> AI must be part of your GSD Prototype phase.
I can understand asking your devs and other employees to try out AI in their workflows in an attempt to get some productivity gains. Sounds like a responsible CEO trying to milk some value out of his existing headcount. But, it sounds absolutely dystopian to tie performance metrics or changing the project planning process to be AI-centric.
I doubt every project fits that paradigm.
Read the post again. Does this look like something a LLM would struggle to write?
A lot of people in upper management spend an awful lot of time writing meaningless word salad. They take a single sentence like "20% is getting fired" and turn it into a four-page essay about "strategic realignment", "streamlining growth", and "enabling the future of the $corp family". It doesn't really mean anything, so there's no way for a LLM to get it wrong.
If you haven't used AI to do complicated tasks where the details actually matter, I'm not surprised you'd get the idea that it is the greatest thing since sliced bread.
Yes, the latter group have tried using it to get work done.
I don't think you can justify the use of "AI-centric" based on his letter. Maybe it's just a projection of fears? But nothing you quoted suggests 50% is more likely than 5%.
whatever happened to "never go full retard"?
talk about being out of touch.
I'm down to try and use AI to enhance processes and encourage people to do that, but making it part of performance reviews is just silly.
This message should be 10 to 20x shorter, to the point and clearly actionable. Instead it feels like we got the output of prompting "can you turn these few bulletpoints into a 3 season telenovella "?
or, as he would say: Clearly the CEO is demonstrating advanced implementation of his own AI-first philosophy by leveraging transformer-based language models to exponentially increase word count while simultaneously decreasing information density—a masterclass in modern corporate communication techniques that validates his thesis about AI integration being not merely optional but fundamentally requisite in today's rapidly evolving digital landscape. The medium is indeed the message.
> AI is now a fundamental expectation at Shopify: Effective use of AI is required for all employees, and it will be integrated into performance and peer reviews. AI is essential for accelerating learning, improving productivity, and advancing business outcomes.
> AI must be integrated into the GSD prototype phase: AI should dominate the prototype phase of projects, enabling faster creation and collaboration. Teams must demonstrate how they can use AI before requesting additional resources.
> Commitment to continuous learning and sharing: Shopify emphasizes self-directed learning of AI tools, sharing knowledge and experiences with the team, and fostering a culture of experimentation to stay ahead in the fast-evolving AI landscape.
I have to question the literacy levels of people these days if a page of text is too much to handle.
This may have been one page, but one full of waffle and slop.
We jumped off the bandwagon this year. AI isn't useful for what we do, and it required too much handholding and reviewing. The only thing it was good for was taking meeting notes, and even there it tripped up on names...despite having access to all the names of the attendees.
LLMs are just the Big Blues of today.
Surely, the work output of most managers, which contains many more natural language artifacts than the typical coders' outputs, should be a better fit for an LLM than code.
I mean, if the job can be done by AI, he should go the extra step and lay off half of the employees, that's the sensible thing to do for any CEO. Why not do that today?
This is pretty dumb. I am looking forward to the news when the CEO is ousted and AI can't even save him.
Ah yes, the 100x engineer delusion. Why not a 1000x?
If the engineer is 10x, the AI is 10x, and they buy a laptop with 10x flops, install lights with 10x the lumens in their office, give them coffee ground 10x finer, they could all easily be 100,000x engineers.
1. It's great to cure "starting trouble"
2. A specific case is to brainstorm ways to layout libraries to handle cases.
3. When I code, using an AI completer is usually faster than looking up docs but with lsp and familiarity with the language, it's not that great.
4. When I'm setting up new software for the first time, i give the AI a few blog posts and books and then have it guide me through while explaining concepts.
5. As a Google substitute to look for things. I get clear answers for most things which aren't too complex.
6. I've used it to explain concepts to me. eg. the difference between ac and dc current.
7. I've used it to create throwaway apps which i don't care about maintaining.
8. I wanted to get a copy a proprietary program to integrate into my application. I couldn't get it. I had an LLM create s dummy version which produces the same output and did my integration. It more or less worked when I substituted with the real tool.
As for the hype, yeah.. Genuinely s problem but there's some meat in there.
And there it is. None of us will be fired or laid off "because AI". The next person simply won't be hired. And then one day, that next person will be you or I.
Really this is a direct, perfect analogy to the industrial revolution. One small person operating a steam shovel could all of the sudden do the work (faster, with higher quality) that for all of human history required hundreds of strong men to do, and it changed the entire world practically overnight.
Businesses simply do not need the mass droves of SWEs who exist to type out CRUD code anymore. We will go the same way of the manufacturing workers of the 20th century, where the small percentage of those who could deeply specialize and adapt and master the processes were ok, but the vast majority never recovered, and ended up in minimum wage misery.
> Businesses simply do not need the mass droves of SWEs who exist to type out CRUD code anymore.
That's true, but not because of generative AI.