Ask HN: What "developer holy war" have you flip-flopped on?
9 points by meowface 22h ago 29 comments
Ask HN: How do you connect with other founders in your city?
5 points by leonagano 1d ago 2 comments
Igor Babuschkin, a co-founder of xAI, has announced his departure
136 TheAlchemist 77 8/13/2025, 11:51:33 PM techcrunch.com ↗
I can only imagine what would cause someone to make such a drastic move. If he's going into Venture Capital then he has no problem with the workload.
Elon seems very talented at finding trends and convincing talented people to help him build the future. He seems equally talented at then pushing these same people away.
Early retirement
The vast majority of advertisers returned, including large ones like Apple plus their expenses were reduced by over half.
They returned in the first few months of the Trump administration when Elon was an important man in the government with direct access to POTUS. But after Musk fell from the Trump's good graces the same advertisers quickly took their marbles and quietly left twitter.
I mean, even if advertisers "returned" (which seems unlikely), it would still be a 3 years of lost growth.
It might also have been done to keep the relationship with Musk, given his apparent political power at the time.
Since I haven't heard a peep of a complaint from Elon's financial backers about having a declining social media site shoved into their AI investment, the investors are still thinking that they could unload on greater fools when the time comes.
You may not care for the owner but the company's tech speaks for itself and the company deserves to be taken seriously based on their production alone.
That's an absurd statement. And one that shows your bias.
> Nobody is choosing Grok unless they want to make spicy/racist meme or show an active support for Elon.
I pay for Grok and Gemini as they are arguably the best two models available right now.
By your logic and your logic alone, this also mean that no one choses OpenAi unless they want to show active support for Sam Altman.
It will never actually matter how SOTA Grok gets (if it ever gets there) when it's output is being guided in that way.
He's an accidental banker with a gimmick: ketamine can make you believe anything!
Politico.eu collected Musk's latest results [1]. Out of what Musk and Trump promised [2], Musk first lowered his promises himself (after getting the money, sorry, the election) from 2000 billion (2 trillion) to 180 billion, a 91% reduction. He didn't make anywhere close to those savings, his organisation then made a further reduction in his promise to 52.8 billion, missing the promised target by 97.36%. But those appear to be inflated claims and actually verified savings only amount to 1.4 billion, 99.93% less than Musk initially promised. BUT ...
Then Trump was going to NOT spend this money. How's that going? Well, in order to NOT spend 1.4 billion dollars, Musk and Trump have spent 7.4 trillion taxpayer dollars. Hey, you missed "saving" by a tiny bit there, Elon.
So Musk, who promised to save the us government 2000 billion dollars, has in fact "saved" the government MINUS 7400 billion dollars. As always, Musk's direct contribution is barely a rounding error.
This number is not counting, of course, what he personally received from the US government which will comfortably put the cost of doing business with Musk for the US government over 8000 billion dollars. I guess that's one way to become "the richest man in the world" (even though I wouldn't trade $1000 of my money for all Tesla shares in existence, so is he really? I mean, maybe $100 for bragging rights, but they don't even come with a certificate anymore)
Let's just say that anyone expecting investments with Mr. Musk to pay off is not just crazy, that is WAY past dangerous insanity. Even his supposed "high-tech" achievements ... well, here's a summary [3].
TLDR: 99.97% miss in performance for Musk's latest project. Total cost to taxpayers of Mr. Musk: over 8000 billion dollars.
[1] https://www.politico.com/news/2025/08/12/trump-doge-contract...
[2] https://www.bbc.com/news/articles/cn4j33klz33o
[3] https://www.youtube.com/watch?v=B71PNEwhyXc
But you know the same story emerges as elsewhere. Mr. Musk's was a financier, nothing else.
There was a time it was reasonable, early space x days for example, when elon was still managable, easily abstracted away and you were able to work with better funding and less red tape vs other options - but modern times i cannot see a reason one would work with elon unless desperate or in the cult, and lets be honest, you dont reason your way into joining cults.
Like most of the "brilliant" people I know that fit the profile you're describing, the usually build product in the same way they maintain relationships, chaotic.
Most people really just DGAF about politics in US. The (few) smart people working for Elon are those who are really into whatever technology and just like to play with toys while getting paid. I used to work with such a dude a few jobs back in my aerospace days, guy was multi talented across both software and hardware, and could easily be making bank at Apple Amazon Google in one of their edge programs, but was content with getting paid like 80k a year in 2010s in a fairly high CoL area with an hour commute all because he got to play with a rather big UAV.
So this is in no way a `no true Scotsman` fallacy. Being able to be well read is a product of having intelligent people in your society, people who make hard things accessible, make it learnable, but it does not itself make you intelligent.
Dont know how much of that has damaged itself post Trump. Seems like Tesla is losing a lot of top talent. Remains to be seen if SpaceX is falling off the wagon.
This is a common fallacy.
If you are truly an intelligent person, and you know what the right solution is, what you want to be is around people who work for you. You don't want to have to spend time convincing other people of your correctness, you want to be able to just tell everyone what to do in detail and have them do it without asking questions, preferably with ability to figure the minutia stuff on their own.
The "A" players team is basically people who see value in working 60 hour weeks. With enough ambition and motivation, stuff can definitely get gone, but its not the same as intelligence. Space X didn't win with reusable launch stage because of the smart people working on it, they won because Elon had money to throw at test over and over again, while almost scamming their supplies in not paying them until way past due. Trying something over and over doesn't take intelligence.
And now, everyone is working on the Starship. While its cool, its absolutely impractical. It would take 8 -20 launches to refuel it in orbit for an actual mission, and the engines are basically on the verge of blowing up when operating due to how precise the system has to be for that high of a pressure ratio.
Personally Im very close friends with someone that worked at Space X early enough to retire at 40 with $2mil+ after stock sale. A mechanical engineer. Yet, I have to help him fix problems with his bicycle of all things, because he can't figure out how a simple bearing interface works (he was tightening an axle that was clamping the frame instead of the bearing inner race due to a missing washer, and as a result his suspension pivot was getting stuck)
A good read on the Talent at Tesla https://www.reddit.com/r/EnoughMuskSpam/comments/99sbwa/form...
When he does achieve it I imagine you will create a new post saying something like "he didn't achieved Starship because of the smart people working on it, but because he had the money to throw at test over and over again".
Regarding the tesla link its crazy to see that post crop up again. I remember reading it the day it was posted. I was one of the first people on /r/realtesla. I genuinely believed that they were toast in 2018. That sub has been proven wrong time and time again and now they have devolved into whatever absurdity as long as it is negative of Musk companies. That post has not aged well at all. Looking back, I suspect a lot of corners were cut when the company was running on fumes. The model 3 rollout was so bad that not even the shorts could have anticipated the nonsense they pulled(for example: the famous tent) but they managed to survive and improve significantly since the early model 3 days.
Like I said in my initial post whether there is still enough talent remaining post-Trump remains to be seen.
A mechanic for example can just memorize a bunch of procedures on how to service a car or a bike, and follow them. An engineer can memorize a bunch of advanced stuff like CAD and Stress analysis, and never really work with his hands.
So in this case, your statement would be correct. But both of these would just be average good, not exceptional, like SpaceX hiring makes it seem.
In general, as a mechanical engineer, your expertise is putting different materials together in certain shapes, matter states, and so on to make them do things. You may not know specifically how a bearing assembly works if you have never worked with it, but you should realize that a bearing is designed to provide rotational freedom to a part, and that mechanically, the thing that is connected to the inner part that rotates should be fully separate from the part that does not.
Funny take because I know multiple talented people who’ve worked SpaceX & Tesla who claim an extreme amount of bloat wrt earnestly negative-value employees in the engineering sectors. This was pre-2020 as well.
Is that why their rockets keep exploding?
The cost is also not fully "turnkey" - if you overwork your employees, fail to pay your suppliers, rely on subsidies, then of course you are going to reduce your cost.
And all of that happened before Musk fully lost his mind.
Efficiency wise from pure physics, they would have been better off developing something similar to Scaled Composites. Air breathing is way more efficient to get to altitude, then you do a High Altitude Orbit insertion, and then do your vertical landing. For stuff like Starlink Satellites that are rather small, you have to do more launches but the cost of launch goes down significantly.
They built the best vehicle that would work, today. Not some distant future of a possible way to get to space. An actual working way to get to space. Built on the backs of NASA engineers of the space coast of FL, Houston TX, CA, and AR. You can armchair architect another way, but they successfully built one. One that is reusable. Over and over and over and over again.
EU then followed. Then China. Then India (ordering could be wrong but). So it’s definitely something that works.
EDIT
Ok, EU doesn’t have a reusable rocket yet.
Im not debating the fact that they were able to put together a product. Im also not debating that it was successful. Nor am I debating that any average person can work on rockets, you do need some engineering knowledge.
Im just arguing that the reason they got there wasn't because they managed to hire the best of the best engineers that are smart in one area but fail to understand how much Musk is doing politically. My claim is that Space X hired enough average intelligence, but super driven people, which is why they managed to build Falcon 9 through act of throwing money at it, and as supporting evidence of this, Starship is currently blowing up despite their supposed "learnings".
However if your stuff keeps exploding after you have had quite a large number of those test events, something is amiss.
>Like the Falcon 9, once the correct inputs are developed, it will become as routine as "Hey Siri..."
TBD. Raptor engines are the equivalent of taking 2 liter inline 4 engine and putting on massive turbos to make it generate 1000 whp. Thats pretty much how you get the high pressure ratio.
e: Ah, flagged and dead. https://news.ycombinator.com/item?id=44897047
They publish something that looks good, and the reality is something else.
https://apnews.com/article/grok-ai-south-africa-64ce5f240061...
These modifications didn't seem to get reflected in the published xAI prompts.
- -
[1] Musk himself, obviously
It is a swindle by definition.