Price the tail
Joined February 2024
- Tweets460
- Following754
- Followers1.1K
- Likes1.5K
Congrats to the RSI team on the model release. 4months to build the entire training pipeline from scratch and deliver this. Probably one of the fastest ramps we’ve seen from a neo lab.
1/ Introducing Flint.
It’s a pretty good model for agentic coding locally.
We @Recursive_SI think a lot of everyday coding will move to models running on your laptop or a server down the hall.
Here’s why we built it.
Btw, this is an updated version of @dawnsongtweets’ original slide, with 2026 benchmarks added
Leo the Bayesian retweeted
Your tweets are so funny dude. They make me laugh so much man. Banger after banger. You should come to my event later this week. Well it's not actually my event, it's my friends event. Okay I was just kidding. It's my friends' friends event. I'm just tagging along. It's going to be sick dude. There are going to be so many AI researchers at this event bro, the scene is fire. You have to be there or else you'll miss the next big data company at the seed bro. Last week, I went to this other party and there was a brand new data company that was incorporated at the start of the party and already was at 100m run rate by the end of the party. It was insane. Dylan Patel led the pre-seed and the ingroup consortium of Elad, Lachy, Sholto, and Dwarkesh led the seed. The round was so hot man. You could feel the heat. Did I get into it? No no. I tried, but the founder wasn't taking angel checks. I'm not actually legally allowed to say his name, before we left the party, they made us all sign NDAs so we wouldn't leak the info to the VCs, but don't worry. I'll tell you, just keep it on the down low. Also just make sure to come to my party later this week, you don't want to miss the next ripper. Well, not my party, my friends' party. Oh actually, maybe let me just make sure that I can bring you as a plus one. You should definitely plan on coming though, everyone from the scene is going to be there. Even the Citrini guy that took the scorpion string straight on his dick. Apparently he had to go to the hospital, but its all good, the antivenom was real. Did you hear about the antivenom that Claude made for the sting? No I don't know anything about it, I just heard about it last week at my friends friends friends party. Just join the scene bro, it's going to be awesome, there's so many events. Every week is tech week when you're in San Francisco. I love it here so much. Want to grab coffee with a YC founder later? I can ask if you can be my plus one
One of the perks of working at OpenAI is getting to play around in infinite token abundancy.
Last month you were solving math problems your professors couldn’t. This month you’re trying to cure cancer.
We’ve acquired Orchestra Bio, a platform that helps biopharma teams coordinate the complex process of drug development.
Together, we’ll build on Rosalind, our AI models and tools for life sciences research, to streamline those operations and help new treatments reach patients sooner.
Welcome @ashokaraj_ and team!
openai.com/index/openai-to-a…
DeepSeek releases literally anything:
>boomer: "They finally gave up on pretraining. Just distilling astra and fable now."
>doomer: "They nuked our token economics. There goes every neo lab’s ARR."
Heard an interesting warning from the head of post trainning at one of China’s top 3 ai labs. Get too fixated on serving economics, and you start letting the infrastructure dictate the pretraining architecture. You risk optimizing your flagship into a flash model: still competitive on the capability–cost Pareto frontier, just at the efficiency end when you were aiming for peak capability.
Heard an interesting warning from the head of post trainning at one of China’s top 3 ai labs. Get too fixated on serving economics, and you start letting the infrastructure dictate the pretraining architecture. You risk optimizing your flagship into a flash model: still competitive on the capability–cost Pareto frontier, just at the efficiency end when you were aiming for peak capability.
Heard DeepSeek is planning to rebrand and drop $100M on the domain that should’ve been theirs all along: openai.com.
Honestly, MIT Han Lab might be the best in the business at this. @songhan_mit is the GOAT.
Song and his students have a knack for finding a very specific MLSys bottleneck that matters enormously to a chipmaker or AI infrastructure provider, then building the algo and systems around it until there’s a product that fits naturally into the buyer’s stack.
From DeePhi getting acquired by Xilinx, to OmniML by NVIDIA, and now Eigen AI by Nebius, you can see the pattern.
which AI labs turn into startups?
counted lab size vs startups founded by members, public sources only
stanford hazy: 7 startups from 13 people
stanford NLP: 12
berkeley sky lab: 7
@chrmanning your group is the only big lab up top
Btw, I’m building Daedalus Research (@DaedalusRsch) with @shawpredicts and other very sharp researchers, working on forecasting models and neo insurance.
If you only know me for the industry gossip and have no idea what we’re building, I’m not building hard enough.
Forward.
Top-tier data vendors frontrun OpenAI and OpenAI frontruns everyone else.
(mom, Nic Carter quoted my post. I made it.)
Erdős problems seem to have overtaken prediction market making bots as the preferred way for larpers to advertise a 150+ IQ. I just wish some of these people actually cared about the mathematics, or had any feel for what makes a proof beautiful.
Was just reading papers from 2006 on risk society and mechanism design.
They go far deeper than almost anything being said about AI alignment and insurance today.
More than five of your data lab’s favorite data labs just liked this post. They seems to have found even more ways to legally commit crime.
This evening I sat across from the CEO of a so-called trading neo lab with over $1B val. He told me their models—built on best-in-class RL environments and online learning algorithms—have produced exceptional pnl and sharpe ratios.
I later learned that the company is, in fact, a data broker LARPing as a frontier lab. Its actual edge is trading stocks and options on insider information of model releases and synthetic-data acquisitions.
It appears I have once again materially underestimated the absurdity of this industry.
In the run-up to 2008, S&P resisted changes to its CDO rating model over fears of losing business. Voluntary AI audits face a similar conflict: when rigorous scrutiny threatens continued access, compromising professional standards can become economically rational.
People worry a lot about third-party auditors being biased because of where they get funding or who their employees are friends with. IMO by far the biggest source of bias is that if they upset an AI company, that company will stop letting them audit.
When I was at METR this dynamic was a way bigger effect than everything else combined. Of course no auditor can avoid this until audits are no longer voluntary.