HN user
Players manage ship AIs to task sub-agents in a Factorio-like space trading environment. The open-source project utilizes Pipecat AI for backend, Supabase, and Vercel for the React frontend.
Happy to talk more about this if you're interested. I have a lot of ideas around this. And see this kind of setup as really important, and benefitial.
I believe this is a fantastic idea, even if only because I literally was ideating on the same lines just last week.
I think there were some usability issues, but overall, solid stuff.
I think if I were you, I would focus on the UX of legislative bill analysis rather than the LLM side of it, but that's just me.
I'm really interested in the reasons behind Greg leaving. IIUC he was doing great engineering related work.
Information also reporting that Greg and others have left, or are about to leave. I wonder how long he's been gone, given he was still on Twitter promoting work until recently...
This was pretty cool to read back in 2021. They also had a previous post in 2018: https://openai.com/index/scaling-kubernetes-to-2500-nodes/ where they had previously hit this limit.
Cool to see infra solutions at OpenAI. I wonder if these are still powering existing solutions. Ben, one of the authors, seems to have left OpenAI.
I wonder if that's true at a certain stage of OpenAI, which because of the product bootstrapping skills of Sam and co, has made his role irrelevant?
I mean, Jakub can take it forward at the current scale and leadership team of Sam and other people, but maybe he could not have earlier, which is where Ilya shone?
Altman's tweet (https://x.com/sama/status/1790518031640347056?s=46) makes it seem as if he wanted to stay, and Ilya disagreed and "chose" to depart. Very interesting framing.
I don't have a horse in this race, and maybe the root comment came off as flippant and disparaging. But I'm not reading "outsiders" as being what you say "gatekeeping".
Maybe another perspective is that "outsiders" may not have the same view of the issue as experts in the field and may not (historically, in OP's experience) seem to want to work together with the experts to develop this view. Handwaving away complexities and not willing to get hands dirty is something I've seen as well so maybe I'm a bit more empathetic, but cold shoulders from "experts" towards newcomers is definitely a thing.
Both of which could help both sides - bring more depth to the fresh view of the "outsiders" and actually bring valuable freshness to the depth of the "experts".
I'm not sure folks who're putting out strong takes based on this have read this paper.
This paper uses GPT-2 transformer scale, on sinusoidal data:
We trained a decoder-only Transformer [7] model of GPT-2 scale implemented in the Jax based machine learning framework, Pax4 with 12 layers, 8 attention heads, and a 256-dimensional embedding space (9.5M parameters) as our base configuration [4].
Building on previous work, we investigate this question in a controlled setting, where we study transformer models trained on sequences of (x,f(x)) pairs rather than natural language.
Nowhere near definitive or conclusive.
Not sure why this is news outside of the Twitter-techno-pseudo-academic-influencer bubble.
This makes a lot of sense at deep tech places that don't sell to enterprises where one can put their head down and just work on technical challenges. A similar place like Mongo DB or the hardware division of NVidia comes to mind. But if you're supposed to iterate with any input from more than 2 people that isn't highly objective (technical metrics that can correlate well with product satisfaction), communication and diffused collaboration is important and necessary. That doesn't mean it has to suck. But the Netflix model you described isn't applicable everywhere and shouldn't be blindly replicated.
Yup, weird outage.
Reloaded the page multiple times and then finally got to the home page.
Was this while using the Apple Watch ?
I'm curious as to what other devices you could use to log your heart rate ?
I just bought "Meditations" by Marcus Aurelius.
Apart from the debate raging on about grabbing/absorbing more from a physical book than an ebook, I feel that I'm simply more attached to the physical aspect of the reading experience.
I've read ebooks when I've been unable to find the hard copies, but I almost always revert to the hard copies.
Well done ! I started reading this - got around to 10 and stopped for a while - and totally forgot about it.
It continued and - finally ! - finished.
PS: Fellow readers, did it ever stop being awesome and/or become unpleasant from the perspective of the original plot ?
I've put this up for people who got selected and want to connect with fellow HNers.
https://news.ycombinator.com/item?id=7623598
Maybe we can do something better.
Mlpack. Multi-Class Adaboost implementation.
I got accepted ! How awesome is that ?
Best ambience-sound site I've come across which for once really works.
The sound of fire and humans is something new I've seen (or heard) in such a site.
What about cricket chirps, and to be a little bold, traffic ? Traffic in the distance ?
It used to be that there was REALLY good stuff posted there. A lot of the initial content there was really helpful, quite insightful. I used to hang out there everyday. The writing experience was clean; you always required a photo and the comment sys was new. It was easily cataloged and a lot of amazing writers hung out there.
That was quite recent; this June, I think.
Then it became popular. Turns out anyone who could write properly, using correct grammar, make it seem like they were thinking differently than the herd, and/or were an entrepreneur. Or want to post some fluffy pseudo-intellectual stuff.
You really have to wade through a lot of bad stuff, to get to the good ones. It started out well. But the section "Editor's Picks" already has like 2900 articles or something.
Nowadays it's more like you select the best from the week's best or the month's best.
Ah. This site. I came across this when I was a kid - like 5-6 years ago - I had googled the term "how to become a hacker". This guy had opened my eyes then - I was 14 then. He still continues to do so. His links are still relevant and an amazing starting point for anyone who wants to learn.
One should probably keep visiting this page like every 6 months or so - just to check whether you are on the right track to learning.
A much needed perspective. All I've read are the H/W specs. Though I'm still apprehensive about spending so much on it, it does change how I see it. Still, better options seem abound.
Arch linux with gnome.