Tim Zaman

GPT 5.2 was fun to work on. We are now so good at training large scale jobs, it's set and forget. Days just go by with the giant cluster humming along. OpenAI has the best and deepest hardware health/workload team out there, it's phenomenal. Team is small team but always hiring!

@dhaval_shroff @elonmusk Beautiful. Put it on Mars Hill though, for the views and valley oaks

Thx @elonmusk

@karpathy Back then, I recall sitting in the nvidia allhands with jensen, tiny crowd, and Jensen says "We are an AI company" and i just shrug: dude this is a gaming company with ai datacenter geeks on the side. I was on the deep learning side and didnt even see it as clearly as he

The signature is alluding to NVIDIA GTC 2015, where Jensen excitedly told an audience of, at the time, mostly gamers and scientific computing professionals that Deep Learning is The Next Big Thing, citing among other examples my PhD thesis (one of the first image captioning systems that coupled image recognition ConvNet to an autoregressive RNN language model, trained end to end). This was back when most people were still unaware and somewhat skeptical but of course - Jensen was 1000% correct, highly prescient and locked in very early.

Thank you Jensen and NVIDIA! She’s a real beauty! I was told I’d be getting a secret gift, with a hint that it requires 20 amps. (So I knew it had to be good). She’ll make for a beautiful, spacious home for my Dobby the House Elf claw, among lots of other tinkering, thank you!!

@MinisterEZK Groningen gas open. Kern centrales. Vooruit met de geit.

De huidige situatie in het Midden-Oosten raakt onze economie. Door prijsstijgingen van olie, gas en energie-intensieve producten, verstoringen in de wereldhandel en toenemende onzekerheid. Het kabinet begrijpt de zorgen hierover van huishoudens en ondernemers. (1/3)

@nikhilville Yes especially iced new manhattan is amaze

@timzaman is new manhattan better than tesora?

Cupertino Philz for your cpus, san jose starbucks for memory

@__tinygrad__ sorry but how do you not have the ip address(es) on the display, it's the killer app for any hw.

it's a red v2 box autoresearching! Claude ported autoresearch to tinygrad. someday soon it will autoautoresearch with a local LLM

@brian_armstrong disclaimer: i bought btc at $11. (..and sold at $15)

@brian_armstrong I think the problem is - they havent been equally distributed across the population. I doubt this will be accepted by the (very very large) out-group.

The 20 millionth Bitcoin was mined yesterday. Now there are only one million new Bitcoins to be mined, which will take over 100 years. Decentralized, inflation-proof, global money.

@adcock_brett Nice work

Today, Figure is showing another major milestone towards a robot in every home Running Helix 02, cleaning a living room fully autonomously

@benjamin_bolte @asimovinc Welcome! Exciting and hope to catch you soon. Didnt know kscale shut down, too bad

I suppose this is as good a time as any to share that I've joined OpenAI. I plan to continue supporting a few open source humanoid companies like @asimovinc and anvil.bot where I can. The next few years are going to be pivotal. Excited to help build the singularity.

OpenAI has gotten really good at training.

@fchollet Every year since 2016 I've said to myself that it's too late to do my ai infra startup. It hard to extrapolate on the peak

If you ever feel like you're late to the game, consider that in the 1890s many scientists thought physics as a field was completely solved (quote below is from Albert Michelson in 1894). On the front of intelligence science, it feels more like the 1870s. For the first time we have something that is starting to really work (however primitive it may be), which we can use as a springboard for the next few decades of discoveries.

Helping AI reach more people requires deep collaboration across the ecosystem. Today we’re announcing new investment, with support from @SoftBank, @NVIDIA, and @Amazon, to scale the infrastructure needed to bring AI to everyone. openai.com/index/scaling-…

@ajassy @OpenAI Looking forward to working together!

Excited about our new strategic partnership with @OpenAI. Developers and companies of all kinds are eager to run services powered by OpenAI models on AWS, and our unique collaboration will provide a stateful runtime environment for them that’s powered by OpenAI’s frontier intelligence on Amazon Bedrock. OpenAI is also going big on our custom Trainium chips, which are 30-40% more price performant than comparable GPUs, to power their growth. Both the leading AI labs now have made significant commitments to Trainium, which is gaining a lot of momentum. We’ll be the exclusive 3P cloud distribution provider for OpenAI Frontier (which enables organizations to build, deploy, and manage teams of AI agents). And finally, we’re excited about our investment in OpenAI — it’s an extremely talented team with great products, IP, and vision for how they can continue to serve customers and enterprises. We think they’ll be one of the big winners in AI, we can help them grow, and we believe we’ll earn a strong return for Amazon over the long term. aboutamazon.com/news/aws/amazo…

This is not due to capability. This is due to people's habits and their influences. When many were using tab complete, others already took the next step (away from handcoding) and knew how to steer the agents more efficiently than tabs. Even now some ppl swear by handcoding. And indeed some are already fluent in parallel codex workflows - which are completely inevitable

Cool chart showing the ratio of Tab complete requests to Agent requests in Cursor. With improving capability, every point in time has an optimal setup that keeps changing and evolving and the community average tracks the point. None -> Tab -> Agent -> Parallel agents -> Agent Teams (?) -> ??? If you're too conservative, you're leaving leverage on the table. If you're too aggressive, you're net creating more chaos than doing useful work. The art of the process is spending 80% of the time getting work done in the setup you're comfortable with and that actually works, and 20% exploration of what might be the next step up even if it doesn't work yet.

We have raised a $110 billion round of funding from Amazon, NVIDIA, and SoftBank. We are grateful for the support from our partners, and have a lot of work to do to bring you the tools you deserve.

@yaroslavvb @karpathy Haha this is so good

Greg and @karpathy attemping to use grill at OpenAI 2017 offsite

@__tinygrad__ @comma_ai $60 for 10Gbps (8Gbps measured) in bay area

We got 2.5G symmetric business fiber to our office for $32/mo in Hong Kong. I'm so upset with what @comma_ai is paying in the US. (speed limited by gigabit card in tinybox red)

@GavinNewsom Before you call shots, get your own books in order. Before then - we don't wanna hear a peep from you, boss.

Donald Trump just announced a NEW 15% TAX on the American people. He does not care about you.

seeing so much positive progress across each part of openai right now, very proud of the team