re-thc

20 ms·

361.
by re-thc 8mo ago
> Like if the users don't even know how LLMs work or what they are good at, why are they being forced to find new ways? Same happened with crypto. Every financial institution was doing or trying to do some blockchain thing. 90% or m
362.
by re-thc 8mo ago
> Is truth a constant or a personal definition! It always has been what you believed in. E.g. at 1 point the Earth was flat. Now it's round. 100s of years later maybe it's a Hexagon. The so-called knowledge and backing all come
363.
by re-thc 8mo ago
> But then you'd get an explanation tailored to your specific perspective Assuming you know how to ask the right questions and make sense of it. Tailored to whose perspective, especially in a team, which then might make it worse.
364.
by re-thc 8mo ago
> Why do we need a second, less detailed and less verifiable copy of the code Because it requires people to know how to read or specify code. Spec driven development is similar to the no-code movement i.e. the specs are closer to functio
365.
by re-thc 8mo ago
> For the last couple of months all the top models have been from the US The commenter you responded said presidency not country. There are multiple parties and possibilities here.
366.
by re-thc 8mo ago
> It makes sense to me that a small group would be much more productive than a large group That's not the scenario. The scenario is a large group vs a large group cut into a small group. The chaos and disruption of slicing 1/2
367.
by re-thc 8mo ago
As in AI is a belief? This isn't a religion. IF things don't add up they don't.
368.
by re-thc 8mo ago
> the correction was largely complete over a year ago We had tariffs and other disruptions since. So more correction is required.
369.
by re-thc 8mo ago
> A canary for this would be whether Gemini skews toward building stuff on GCP Sure it doesn't prefer THE Borg?
370.
by re-thc 8mo ago
> but something has changed i.e. we finally decided to audit head count from post covid-era. > paired with smaller and flatter teams i.e. management was axed
371.
by re-thc 8mo ago
Not that 1. Wink.
372.
by re-thc 8mo ago
> It's not awesome, not for us. Depends on where you stand. Maybe leet code won't be a common thing (can be solved with AI), maybe they'll look for different skills, etc. If losing 30% means hiring the right people for the
373.
by re-thc 8mo ago
Product Hunt (I assume)
374.
by re-thc 8mo ago
> But a short line "AGI is possible, powerful and perilous" > At which point the question becomes: is it them who are deluded, or is it you? No one. It is always "possible". Ask me 20 years ago after watching a sci
375.
by re-thc 8mo ago
> I think OpenAI has better chance to winning on the consumer side than everyone else. Which doesn't make money. > Of course, would that much up against hundreds of billions of dollars in capex remains to be seen. Most of that is
376.
by re-thc 8mo ago
> Apple, for example, designs nearly the entire SoC in-house I already mentioned Apple lost their top spot. They're also are paying a 100% price hike for Samsung RAM. > Spending billions to help Broadcom get into par with Nvidia
377.
by re-thc 8mo ago
Wasn't OpenAI's moat buying up all the RAM or Nvidia cards?
378.
by re-thc 8mo ago
What!? I always thought we could just download more RAM!
379.
by re-thc 8mo ago
China doesn't need to buy it. They can continue their policy and look good. They've already found a better route. Buy it elsewhere e.g. in Singapore. Train their models there using Nvidia hardware. Ship the result and fine tune ba
380.
by re-thc 8mo ago
Google doesn't and won't have enough TPUs regardless. Nvidia owns the supply chain (confirmed with them trying to or has taken over Apple as TSMC's biggest customer).
381.
by re-thc 8mo ago
It's outsourced to Codex
382.
by re-thc 8mo ago
> and about the same as Microsoft > Must be nice to pull numbers out of one's ass with zero consequence. Seems accurate? What they are saying is if Microsoft ends up buying the rest of their shares then i.e. Microsoft's tota
383.
by re-thc 8mo ago
It's in the title. Software 3.1 (years ago).
384.
by re-thc 8mo ago
> run the AI function once it would be much simpler to cache the resulting code than repeatedly re-generate it? Surely, you'll run a function that does an AI call to cache the resulting code.
385.
by re-thc 8mo ago
> Postegres is all the rage, MySql is the og From "influencers" i.e. YouTubers? OG real world was Oracle and SQL Server.
386.
by re-thc 8mo ago
> puts them a little closer to DigitalOcean's pricing DigitalOcean has been losing out to hardware inflation for a long time. Most of their fleet is very old except maybe the very top tier and that's still old. They're rea
387.
by re-thc 8mo ago
RAM had been sold out so…
388.
by re-thc 8mo ago
Does that mean they will upgrade the fleet? Can we get AMD Turin across the board? US cloud has been lacking.
389.
by re-thc 8mo ago
The free demand is increasing. With price increases hitting the cloud, will there still be demand?
390.
by re-thc 8mo ago
Planned doesn’t mean achievable. The yields are still low.

More ›