ElMeanYo

[Chief_R @Chief_R_]

Not my usual post topic........
But I’m sick of the Boomer hate. Everywhere I look, it’s the same thing. Boomers ruined the economy. Boomers hoarded all the houses. Boomers wrecked the planet. Boomers made it impossible for anyone younger to get ahead. Like my entire generation sat down one day and decided to screw everyone who came after us. I worked. I saved. I paid into the system for decades. Now I want to spend some of what I earned, and suddenly that’s a moral failing. Wanting to live in the house I paid for is “hoarding.” Taking a vacation is “stealing the future.” Existing past 65 is apparently an offense. Housing is expensive. Wages stalled. Costs went up. Fine. That’s real. But acting like every Boomer personally priced you out of a starter home is lazy. Governments printed money. Corporations consolidated. Zoning locked cities down. Colleges sold a $200k degree for a $50k job. Interest rates, regulation, immigration, supply, demand — pick a cause. “Old people exist” is not an analysis. And the entitlement in it is wild. Nobody owes you the same deal your parents or grandparents had. Life doesn’t come with a guaranteed starter home, cheap college, and a pension. Plenty of people in my generation got crushed, too. Plenty of younger people are doing fine. Outcomes aren’t a morality play by birth year. I’m not apologizing for working, saving, and wanting to enjoy the result. If that’s “Boomer hate” fuel, so be it. I’m done pretending I should feel guilty for not being broke.

Rant over.
Thanks for coming to my Ted Talk.

[ElMeanYo @ElMeanYo]

@AlexBores They were told to hack and they hacked. So what?

[ElMeanYo @ElMeanYo]

@ashutoshrana_20 Tests, tests and more tests. Integration tests, performance tests, business logic tests, ui tests. Then make sure you stand those tests and what they are measuring. If you have a thorough enough test suite, it doesn’t matter what the code looks like.

[ElMeanYo @ElMeanYo]

@larrykim Your software just isn’t that complicated then

[ElMeanYo @ElMeanYo]

@HowToPrompt__ No. The boring junior level work will just be bumped up several levels.

[ElMeanYo @ElMeanYo]

@OskarWieckowicz It’s been completely opposite. Claude has made me busier than ever.

[ElMeanYo @ElMeanYo]

@donnfelker I can do this with just Claude code and my iPhone now, without all this setup

[ElMeanYo @ElMeanYo]

@dimitrov2k You’re still looking at code?

[ElMeanYo @ElMeanYo]

@Oilfield_Rando So then sell it and laugh all the way to the bank. What’s the problem here?

[ElMeanYo @ElMeanYo]

@StevenShortino @SMB_Attorney Why scary? Finally housing is becoming affordable. That’s good news.

[ElMeanYo @ElMeanYo]

@_The_Prophet__ So 4 years ago we were complaining housing is too expensive, now we’re complaining it’s becoming to cheap?

[ElMeanYo @ElMeanYo]

@NBAMemes Incredibly hard to throw a ball of the rim and have it come right back to you. Not to mention catch and pass in midair like that.

[ElMeanYo @ElMeanYo]

@japan_nobunaga Just not true:

https://t.co/T7UByYpYIK

[ElMeanYo @ElMeanYo]

@t_sadiity Having weekends free is only a recent thing in human history

[ElMeanYo @ElMeanYo]

@mitchellvii @behappy5467 Use Grok for current events Bill. Different AIs have different uses and Grok being tied to X usually has the latest info.

[ElMeanYo @ElMeanYo]

@ashleypeacock No, because I don’t have to manually type code anymore

[ElMeanYo @ElMeanYo]

@TheInternetFish

[ElMeanYo @ElMeanYo]

@cremieuxrecueil How much of this is credit card debt?

[Kun Chen @kunchenguid]

opus 5 is a sign that the obsession with “long-horizon agents” in model training is finally backfiring

i don’t like long-horizon agents, and i’ll explain why they fundamentally don’t work

some people will immediately jump out and say “skill issue”. well, show me one profitable business you built with a long-horizon agent working all by itself - i’d love to learn

so far, the only thing they were able to build that’s even interesting enough for people to talk about are those 3d games that are a partial clone of something that already existed

the reason an agent was able to build a working prototype of complex games like call of duty was that a team of humans already figured out all the requirements years ago for how such games should work, what kind of controls are intuitive, what mechanics are fun etc

all those requirements were already absorbed into the model weights, so when you say “build me call of duty” the model already knows the details. its long horizon execution capability can get all the requirements implemented, which i must say is indeed impressive

but now you can see - the value of long horizon execution has a prerequisite of a massive amount of high quality requirements clearly defined upfront. it took a big team of very talented humans months of effort and many iterations to define that for call of duty

now imagine games like call of duty don’t exist yet, how would we use agents to build it for the first time? we can’t say “build me call of duty” any more. and there’s no way we can define months-worth of game design details upfront

we’ll have to build a tiny prototype of the most basic mechanics, play with it, see if it’s fun, then iterate and expand the complexity. even with the smartest humans, that’s how we work towards something great

we don’t need agents to go dark for a long time, spend tens of thousands of dollars worth of tokens, and come back with a product the agent randomly decided to build - try build something truly novel with this and you’ll see it can’t come up with anything that’s actually profitable (i’ll show you why in a bit)

we need a tight feedback loop where we can collaborate with the agent, plan with it, understand what it’s done, question its approach, apply our judgement, give it real world feedback and iteratively arrive at a good outcome

and that’s exactly what opus 5 absolutely suck at. why? i explained it in more depth with my previous post on how RLVR works - RLVR trains the model to generate code that can pass predefined tests in an isolated environment, which is fundamentally incompatible with the idea of having human in the loop. the more we train the models with RLVR to be “long-horizon”, the less they care about talking to humans

ok now - why do they have to talk to humans? why can’t the models iterate and apply judgement by itself?

maybe one day they could, but not today, due to many limitations. two examples -

1. LLMs today can’t “watch a video” yet. they can look through a lot of screenshots, which is extremely inefficient at observing a high fps animated signal. so anything that requires continuous visual attention is something LLMs can’t do very well

2. LLMs don’t truly understand what’s “intuitive” or “pleasant” for humans. they know what’s already proven to be intuitive and pleasant in the past, but if you present a truly novel concept, it can’t predict whether humans will like it accurately

because of those limitations, human judgment is still needed for almost anything valuable. without humans in the loop, agents will only be able to repeat something that already existed, or go in random directions without true understanding of whether it’s building something useful

in summary, long horizon agents assume requirements all exist upfront. they are fundamentally against human in the loop. and they don’t have true judgement for what humans like

that, my friend, is why they don’t work

[ElMeanYo @ElMeanYo]

@AndrewCurran_ Claude: make me some paperclips