Hacker News .hnnew | past | comments | ask | show | jobs | submit | FartyMcFarter's commentslogin

I may have missed this in the article, but:

What was the net effect of the optimisations? How much faster did it get?


The mindset of law breaking probably carries across jurisdictions.

It’s all about professionalism.

As long as there's an RPC connection established and a partially sent request, I think it would count.

If any company announces that they use token consumption as an employee performance signal, for me that's close to a red flag to stay away from that company.

No company with good engineering leadership should act like this is remotely a good idea.


As I have snarkily observed at work: if I go $100 over the meal allowance on my business trip, I'll have to have an unpleasant conversation with my manager or finance. If I use $500 in AI tokens unproductively I'll be recognized for being a top AI adopter.

I have seen this type of behavior happen many times in different companies.

For example, at more than one company I've worked for, if you wrote shitty code but got it into "testing" faster than anybody else, you are considered a superior programmer. And then, if you fixed the hundreds of bugs found in your code seen as an extraordinary programmer going above and beyond the call of duty.

Management is always measuring the wrong thing.


You'd be surprised, I know a few devs in very big tech companies, not faang but you definitely know them, and they all have some kind of token leaderboards, a few told their dev "we don't want you to write a single line of code manually anymore", etc.

I assume the execs perspective is something like: if the top 20% of worker produce 80% of the code with LLMs and the company still works then we can get rid of the bottom 80% of devs and save money


But even if the end goal was to lay off 80% of programmers, shouldn't the 20% to keep be the developers delivering the 80% of the code, regardless of whether they spent the most to do it? Like what if the 20% of workers spending the most tokens were actually the bottom 20% in terms of delivery because they were using the worst prompts and having AI constantly implement 5 different versions of everything, then throwing it all out because their prompt was so bad anyway?

Ah, but "who uses the most tokens" is a number, a number generated by a computer no less. Questions like who delivers lots of high quality work require you to do research and make judgements, which is work.

Problem is that those 20% depend on the code reviews of the 80% for some form of pushback.

I think there's probably something to token use as some kind of metric. If you aren't using these tools much, you're definitely not going to remain a top contributor. The world is evolving quickly here.

But it's just one signal out of many, and more isn't somehow inherently better beyond a certain point.


Tokens are the new "lines of code per engineer". Easy to graph, easy to "manage".

The new TPS reports!

Oh, so that was actually a Token Per Second report! Wild!

...and easier to bill! Back, then noboday had the idea to charge per "lines of code", but today it seems accepted to charge per words processed?

Meta does this. Guess what one of the criteria for their recent layoffs was.

Meta tracks token consumption, but has explicitly stated that it is not a primary performance metric. Instead, employees are evaluated on "impact."

Indeed, they also said that previous time off for ill health wasn't a reason either.

but looking at the number of people who had taken leave, it suggests otherwise.


You can believe whatever conspiracy theories you want, of course, but the most straightforward explanation is that when you lay off X,000 or XX,000 people, some number of them will be on leave.

My understanding is that they must layoff some workers on leave, otherwise they would be discriminating

In Europe people on sick leave are protected (under certain conditions).

In London, large number of people on sick leave, or were recently on sick leave were fired. I was there, I saw the people, did the numbers.

If you did the numbers, would you share them with the rest of us instead of just alleging secret facts?

Megacorps being dishonest and human-hostile under capitalism is hardly a theory.

And my CTO insists that PR count isn't a performance metric. But guess what number gets used the minute people are forced to stack rank (of course they don't call it that, but... that's basically what it is)?

You get what you measure.


Famously honest and on-the-level company Meta, who we can parrot the word of uncritically and unquestioningly.

Sure, and I have a bridge to sell you. Or alternately refer you to the inevitability of Goodhart's law.

Do you have any source for this at all? I’ve seen so many different exonerations for Meta’s layoff criteria including claims that engineers using the most AI were laid off because Meta had them build AI tools to replace themselves.

Everyone is oddly confident despite all of the conflicting explanations.


Without any evidence, I would be shocked if performance rating wasn't a factor in the layoffs. But performance rating is not the same thing as AI tool use.

I worked at Uber from 2022-2025. The engineering culture was pretty abysmal, so it checks out.

The problem is that many companies which had reasonable leadership in the past with the advent of LLM AI started to make rushed (and dubious from my point of view) decisions - using token usage to evaluate an employee performance is just one of them.

I worked at a YC company that was doing this and left last month. I wonder where this all started from, VCs and tech execs are such a monoculture

https://x.com/garrytan/status/2018368316033946021

"I am currently averaging about 10k LOC per day (35% of the lines are tests) so wow, 15k/day is #goals"

Garry Tan - YC CEO and LLM psychosis victim (but he doesn't realise that yet)


The where may be the decision makers chasing social media trends. A friend sent me a link to this, this morning, about devs rather than managers, but I suspect it's the same: https://youtu.be/IW3Sbe0Hbgg

Everything about this story is so satisfying, that if I read it in a lesser source I would be doubting it.

The person finding the baby was the person who eventually adopted him. The judge asking the guy to adopt the baby was the same judge that performed the wedding of the couple doing the adoption. Just so many great details.


There is an excellent children’s book about their story. It’s hard to read to my kids without getting teary, tbh.

https://www.petermercurio.com/our-subway-baby/


Many of us could easily imagine, once chance puts us in the position of the person who can’t walk past, taking a concerned interest in the outcome and, realising that the baby would go into the care system, stepping up and doing the massive step of talking it upon ourselves to provide that home?

Imagine, perhaps. Stepping up and doing it, that's more difficult. Then consider this scenario: two young men taking in a child when they did not have financial or social security. I'm not sure what the situation was like for gay men in NYC at the time, but it would be years until they could legally marry. Heck, they weren't even living together at the time. That takes a whole lot of bravery.

(I will acknowledge that it is not impossible to imagine. I have known people who have adopted the children of strangers after reporting their family for abuse.)


> I have known people who have adopted the children of strangers after reporting their family for abuse.

“When I was a boy and I would see scary things in the news, my mother would say to me, "Look for the helpers. You will always find people who are helping.” ― Fred Rogers


[flagged]


God bless the woman who felt they had no other choice & give them help rather than punishment. We will never know the circumstances but we can obviously assume they weren't very much better than the baby's to have done this. Nobody doing something like this out of disdain for the child would have carried it to delivery (regardless of their stance on abortion).

At least in NY State, babies up to 30 days old can be left for care by the state.

> New York State's Abandoned Infant Protection Act allows a parent to abandon a newborn baby up to 30 days of age anonymously and without fear of prosecution, as long as the baby is abandoned in a safe manner.

> A parent is not guilty of a crime if the infant is left with an appropriate person or in a suitable location and the parent promptly notifies an appropriate person of the infant's location. A hospital or a staffed police or fire station are examples of safe and suitable choices.

> A person leaving an infant under this law is not required to give his or her name.


I think every state has a similar type of safe haven law these days. This one for NY first came out the same summer as the story in the article starts. This and similar stories were making a lot of news around the time and people were really interested in trying to help. It doesn't make things perfect, there are e.g. still occasional situations where a person isn't in their own control or in a sane state, but these kinds of changes did help thousands of cases since.

I think a lot of school sex ed programs/health classes have started teaching this to kids as well. Not all of course, as with most things sex ed.


Had you considered this was not in the mother’s control, was not her choice, and/or that this was a better outcome for the child and she knew that?

He was not left in a bin or dumpster. This is not an ideal way to give up a baby for adoption, but don’t assume he was unwanted, or unloved: you don’t know - or seem able to imagine - the full story.


Personally I would be more worried that the woman that had just given birth received the appropriate medical attention and help with whatever circumstances made her feel that she had to abandon her child.

In response to a heartwarming story you did not read, you made an account on this website to advocate for harshly punishing a child or young woman acting under extreme duress for a crime committed 26 years ago?

Quite often in these cases the mother is an abused minor themselves.

What are those safe ways?


In the US anyway there are places you can drop off a child "no questions asked". Often fire departments or (I think?) police stations. It's to prevent this exact thing.

Where does it say the baby was in a dumpster or bin?


Yeah, and back then people moved the goal posts too, saying Deep Blue was just "brute-forcing" chess (which isn't even true since it's not a pure minimax search).

Deep Blue was brute forcing chess in the sense that AlphaGo wasn't brute forcing Go.

Both of them contained a search algorithm that explored some moves from each considered position, usually not all moves. Both of them contained logic (learned or programmed) to evaluate moves and/or positions.

The differences between them are many, but brute force doesn't enter into it in either case.


> One in three employers spent more on restaffing than they saved from the original layoffs. That is not efficiency. That is a wire transfer with extra steps.

Well written :)


It seems this particular company makes a payment for completing those tasks, so it might not be that bad.


We do, it's a part of our hiring pipeline: https://archestra.ai/careers


As someone without strong feelings on Linux vs Windows (I've used and developed on both about equally): this kind of news, along the way Windows has been changing has me wondering if I should change my primary desktop environment at home to Linux.

In my eyes, Windows used to be the desktop environment that "just works and can run almost everything". Lately it's becoming enshittified, with weird bugs showing up more and more frequently (a memorable one is not being able to launch Notepad from the start menu!!). I think Microsoft is losing its best attributes when it comes to consumer software. Linux may not be perfect but it's looking more and more attractive in comparison, even with its imperfections.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: