jfaat

5 ms·

1.
Nvidia Announces RTX Pro 5500 Blackwell Workstation Edition(nvidia.com)
3 points by jfaat 3d ago|1 comments
2.
by jfaat 4d ago
Probably
3.
by jfaat 1mo ago
> there are ordinary looking trucks puttering about on Taiwan's highways with likely tens of millions of dollars of chip inventory sitting inside. Like what happens if one of them gets into a crash? Will Nvidia miss earnings because
4.
by jfaat 1mo ago
By lighting VC (public soon) money on fire...
5.
by jfaat 1mo ago
That's fascinating, it's WAY better than luna ime. What sort of things are you testing it for?
6.
by jfaat 1mo ago
> truly impressive growth hacking on children I haven't been more viscerally disgusted by a phrase I've read about tech in a while. Astonishing word choice.
7.
by jfaat 2mo ago
Your napkin math only shows that the market is identical. There's about 35m businesses operating in the US. So by selling to 10% of those you sell the exact same number of robots...
8.
by jfaat 2mo ago
In the US (parent mentioned US specifically) I think that's just Frontier now that Spirit is gone. I mean technically that's doable sure but idk if I would say trivial it's really limited on routes and the experience is terri
9.
by jfaat 2mo ago
I disagree. The point of the frontier models is to do everything as well as possible ("AGI" race or whatever) but smaller models with some RL are going to be the clear winner for a ton of use cases. Think about all the use cases f
10.
by jfaat 2mo ago
Is your point that that makes this better?
11.
by jfaat 2mo ago
Ha, Konami code and all! That's impressive. I think Kimi has had the best design sense of any frontier-ish model for about 6 months.
12.
by jfaat 2mo ago
Not to mention the bill you would be paying anthropic could be way higher at that scale
13.
by jfaat 2mo ago
The extent of my expertise here is what I saw in a 30s video. I guess the point is it drifts while pointed in the right direction. That could be closer or further from the goal depending on wind and current but keeps you on track at least.
14.
by jfaat 2mo ago
I saw some of her videos on instagram during the journey. There's a rudder that keeps the boat pointed in the right direction without propulsion while she sleeps.
15.
by jfaat 2mo ago
That's why I'm switching to open weight models. I'm running models locally, self-hosting and using third party for various things. Like it says in the article each model has its strengths and none of the proprietary harnesses
16.
by jfaat 3mo ago
Yeah this is sounds close to my workflow and its good to hear you've find a nice flow too! It frees me up to spend that effort on doing more things in parallel and focusing way more on the specs which is usually a good idea anyway.
17.
by jfaat 3mo ago
My whole point is that I don't want it to build an entire feature from one prompt. At most, I want to work with an agent to nail down the spec and then work with an agent that orchestrates the implementation via other agents, same for
18.
by jfaat 3mo ago
We built our own and aren't done open sourcing it but before that I got to a really good place with opencode plus some custom agents, pi family is good too although I haven't used it as much. We made an agent to design a spec, one
19.
by jfaat 3mo ago
> but if you only want to use the best model available, it isn't there yet I'm trying to wrap my head around exactly why so may people seem to want the best model available when it has recently become clear that most halfway de
20.
by jfaat 3mo ago
Known to who? I just did a quick search didn't see anything that like a scientist or physician said about this. Someone who describes himself as a 'biohacking educator' said it without citing sources though.
21.
by jfaat 3mo ago
I'd call 'more likely' an extremely safe take given that it's exactly what's happening right now
22.
by jfaat 4mo ago
I'm going to start using malapropisms so people know I didn't use an llm to write things
23.
by jfaat 7mo ago
Which of the countries that the US has recently attacked are you comparing to Nazi Germany?
24.
by jfaat 7mo ago
With what's happening in the US post covid, I'm gonna have to disagree
25.
by jfaat 7mo ago
Yeah that's a good idea. I played around with kimi2.5/gemini in a similar way and it's solid for the price. It would be pretty easy to build some skills out and delegate heavy lifting to better models without managing it your
26.
by jfaat 7mo ago
Lol wat? I mean you certainly have enough control self hosting the model to not let it join some moltbot network... or what exactly are you saying would happen?
27.
by jfaat 8mo ago
I'm using an openAI realtime voice with livekit, and they said they have a livekit integration so it would probably be doable that way. I haven't used video in livekit though and I don't know how the plugins are setup for it
28.
by jfaat 8mo ago
I hadn't considered that name specifically but I'm not married to the branding! I appreciate that feedback and your other comment validating I'm not the only person with this problem. Happy to chat more via email (in bio)
29.
by jfaat 8mo ago
I see some chess players so I want to plug the chess coaching app [0] I'm building. I don't know many chess players and could use feedback, but I had been paying for chess.com premium and tried some others and it's always gam
30.
by jfaat 8mo ago
I have a lot more to optimize before I'm crunching down the positions but I just made a chess platform[0] with the intention of tracking your play style over many games (integrated with chess.com only for now) because the other ones I&

More ›