uptownfunk

So what is tslas new story going to be at earnings.

This is insane. We are wired so strangely.

I have gone through something similar where I won’t feel motivated to even pursue the interview properly because I’ve been burnt so many times already. Yes you just need one yes (ideally more for negotiating leverage) but those 1-2 offers are damn hard to come by these days

Absolutely brutal, difficult.

My observations for the last two years:

Employers are hyper hyper selective

There are a lot a lot of fake job postings

The bar has gone way way up

It is in general very difficult and gruelling to get a job

If engineering roles are hard to find, product is 10-100x worse

If you have a job, be thankful, don't rock the ship

OBVIOUSLY there will be statements to the contrary, but this is really more in the spirit of - if you're having a hard time out there, you are not alone, and at least one person out there (who has been working for the last 10 years or so) finds it much much harder. I would say reminscent closer to 2014-2015 than anything like 2020-2021.

For founders it’s great. You don’t need a ton of startup capital to actually start something. Easier than ever. I think this means there will be less of a supply of goood early stage deals. And if VC can’t get in on the ground floor it becomes harder to get the multiples they sell to their investor. It also means that the good companies will be able to get to finance their own growth which means that whatever is left in the later stage just won’t be as compelling. As an LP you know have to look at potentially other asset classes or find investors who are going to adapt to still find success. Otherwise I think AI is most disruptive here now to the VC and by implication their capital markets. I think late stage financing and ipo market should not change substantially.

Can you share the playlists?

For me they provide a good jumping off point. I do find I still need to read it and tweak it. But it’s much easier than having to write the damn thing myself. I am not trying to be an engineer, I am trying to build a business and get an mvp off the ground. And if we get anywhere, then maybe we can have budget for real engineers.

LLM (chatGPT) has mostly replaced 90% of Google search for me at this point. (Except browser, email, image, YouTube, google docs)

That is a valid constraint. For us we don’t really mind if it emits anything sensitive we are more concerned that it is correct.

Not talking about you!

PSA: It’s really rude when someone does a show HN to go and plug your own competitor. Let them have their moment

In ancient times Hindus had a prayer mat made of kuusha or kaasha grass. Apparently the grass descended from the hairs of the boar incarnation. So this notion of a seat in worship is indeed very ancient.

AWS is just playing copycat with msft. They rarely have any good original ideas. Other than IaaS and online retail.

Cost of inference will tend to the the same as cost of a Google search. It is infra that will come down to negligible and almost free. Then as others have said it will tend to freemium (pay to have no ads). And additional value added services as they continue to evolve up the food chain (ai powered sales, marketing, etc)

Yes, that is the case. It is largely 4B in capex investment, I’d imagine 10% or less is cash. One would think nvidia could get much better terms investing its gpu (assuming they can get it into a working cluster). Instead it’s nvidia gets cash for gpu hardware, that hardware gets put into a data center and AWS invests their hardware as credits for equity instead of cash. And because AWS has already built out their data center infra they can get a better deal than nvidia making the play because nvidia has to rebuild an entire data center infra from scratch (in addition to designing gpu etc).

Now if AWS or gcp can crack gpu compute better than nvidia for training and hosting, then they can basically cut out nvidia and so essentially they get gpu at cost (vs whatever markup they pay to nvidia).

Because essentially whatever return AWS will make from Anthropic will be modulated by the premiums paid to nvidia to invest and also the cost of operating a data center for Anthropic.

But thankfully all of that gets mediated on paper because valuation is more speculative than the returns on nvidia hardware (which will be known to the cent by AWS given its some math of hourly rate and utilization which they have a good idea of)

What are the bottlenecks that make this so slow

If they could just make a dplyr for py it would be so awesome. But sadly I don’t think the python language semantics will support such a tool. It all comes down to managing the namespace I guess

All your corporate culture, comp Structure, Interview process etc etc is all so much meta if you can’t deliver. They showed they can’t deliver. Huge let down.

Hahaha this reads very close to what actually happened

Possible but unlikely given how much demand there is and the pressure to deliver returns to shareholders, however sure it is possible. Right now search is very inefficient, the search space is massive. That is the main problem. You can have many sequences of text that sound plausible, but of them a much smaller number will be logically valid. This is the main challenge. Once we can search efficiently not just in semantically valid space but I suppose what you can call syntactically valid space then we will be able to crack this.

Ok but how do you get around needing a 10k or 100k h100 cluster

The hard part is in the creation of new math to solve these problems not in the use of existing mathematics. So new objects (groups rings fields) etc have to be theorized, their properties understood, and then that new machinery used to crack the existing problems. I think we will get to a place (around 5 years) where AI will be able to solve these problems and create these new objects. I don’t think it’s one of technology I think it’s more financial. Meaning, there isn’t much money to be made doing this (try and justify it for yourself) and so the lack of focus here. I think this is a red herring and there is a gold mine in there some where but it will likely take someone with a lot of cash to fund it out of passion (Vlad Tenev / Harmonic, or Zuck and Meta AI, or the Google / AlphaProof guys) but in the big tech world, they are just a minnow project in a sea of competing initiatives. And so that leaves us at the mercy of open research, which if it is a compute bound problem, is one that may take 10-20 years to crack. I hope I see a solution to RH in my lifetime (and in language that I can understand)

Wow it’s really bad.

What is this used for?

It was so bad. So so bad. Like don’t use your customers as guinea pigs for live streaming. So lame. They need a new head of content delivery. You can’t charge customers like that and market a massive event and your tech is worse than what we had from live broadcast tv.

ARC is a bigger contribution than Keras. It’s great he had two major contributions. I can’t wait to see how they crack ARC

How many are there that we don’t even know they are targeting. That is what would keep me up at night

He is one of those for whom I wish I had enough time to understand his ideas.

I miss this era of life and childhood