Labib Tazwar Rahman

Announcing our $35 Billion Project Concord 🇺🇸🇰🇷 Excited to unveil Project Concord with our first project: a $35 billion “Living Data Center” in South Korea — where AI designs, optimizes, and orchestrates the full stack, from silicon to systems to the entire site. Voltai is partnering with Stock Farm Road, an investor group founded by Brian Koo, the grandson of LG founder Koo In-hwoi, and Amin Badr-El-Din, founder of UAE Offsets Group, which spawned Mubadala, UAE’s sovereign wealth fund. Like Voltai, it also counts Alphabet Inc. (parent company of Google) Chairman John Hennessy as an advisor, alongside former Vice-President of the European Commission and Salesforce Board Member Neelie Kroes. Voltai and Stock Farm Road are the primary partners of Project Concord. This initiative aligns with South Korea's push to scale AI infrastructure. President Lee Jae Myung just announced plans to triple government AI spending in next year's budget. Project Concord marks the first time AI will serve as architect, manager, and operator of AI infrastructure that is fully vertically integrated, from silicon to rack to the whole facility. We're building the playbook for how AI infrastructure scales for the future of AI. News covered by The Wall Street Journal

Pax Silica, one of the finest initiatives to drive multilateral alignment on securing the silicon supply chain, thanks to @jacobhelberg Great to witness here in DC.

@vsikka Congrats!

Hello World! I'm excited to share the launch of Hang Ten Systems, a new endeavor to help enterprises thrive in the age of AI. AI is upon us all like a massive new wave. And I learned a long time ago that when there are big waves around, it is time to surf. Not just to surf, but to hang ten — to master the wave so well that you can walk all the way to the front of the board and hang your ten toes off the front. Hang Ten is already helping some of the world's biggest and most important enterprises — like Fresenius, Siemens Energy and others — hang ten on the biggest wave of our lifetimes. Our dream is to help enterprises not just transform with AI, but use it as a force to do what no one could do before. We're backed by a remarkable group. Mayfield leads our round; @NavinChaddha and I were students together at @Stanford and always looked for an excuse to work together. They are joined by @aramcoventures, the strategic venturing arm of Saudi @Aramco, one of the world's largest companies and a key leader in energy and infrastructure, as well as some of Silicon Valley's best-known angels. And I'm privileged that Jerry Yang — also a friend since Stanford — serves on our board. Building Hang Ten with me is the core team I've worked alongside for years: @navinb, @sanjaypaloalto, Tao Liu, Frank Yu, Pradeep Panicker, Yusuf Safdari and ten other big wave surfers. Over time I'll share more about Hang Ten and our work. For now I'll say only this: I have seen, firsthand, the dramatic things AI delivers for the people and teams who somehow just know what to do with it — I have watched them, and myself, reach in minutes what could take teams years of toil. And I have seen the far greater number who get none of it, and who often end up causing harm instead. In that gap lies the biggest opportunity of our time. It is time to ride this wave. If you're a surfrider — someone who lights up at the chance to help businesses solve the hardest problems they face — write to us (jobs@hangten.ai)! — Vishal

@jubayer_hamid awesome blossom @jubayer_hamid and @ifdita_hasan

The most capable reasoning systems in AI scale inference compute along several axes: sequential compute to think longer, parallel compute to sample many independent attempts, and aggregative compute to synthesize prior traces into a new improved one. But during training, we only optimize how models use sequential compute. This creates a fundamental mismatch between how we ultimately deploy these systems and how we train them, leaving much of search and synthesis unoptimized. We introduce SPIRAL, an RL framework for making all inference-compute primitives end-to-end learnable: models learn to coordinate sequential, parallel, and aggregative reasoning using only the reward of the final output. Work with @ifdita_hasan (co-lead), @michaelyli_ , @oshaikh13 , @yoonholeee , @DorsaSadigh , @chelseabfinn , @noahdgoodman 🧵

I do not yet know what VOLTAI is or does but bless them.

VOLTAI chartered a yacht for their happy hour at Reindustrialize. This is how you do it.

Designing a single advanced chip can take years, thousands of engineers and billions of dollars. Voltai is building AI to change that. Congratulations to Voltai on being named one of the @wef 2026 Technology Pioneers, a cohort of 100 early-stage companies selected worldwide for their work at the frontier of technology across AI, energy, space, and biotechnology. Founded out of a @Stanford AI research lab, Voltai develops AI models and agents that learn to design semiconductors and electronics, reasoning across hardware, software and manufacturing to push design beyond human limits. It is a hard, foundational problem that sits beneath almost everything the AI era depends on, and one that will be explored under the Foundation Track at @RaiseSummit . This recognition is well earned, and we are proud to have Voltai joining us as an Official Partner at RAISE Summit this July 8-9. 💡 Learn more about the 100 companies selected for the 2026 Technology Pioneers community: weforum.org/stories/2026/0…

@zacharylipton @thinkymachines @OpenAI you’re my favorite shitposter on x this weekend

BREAKING NEWS: Barret Zoph rejoins @thinkymachines as Chief Financial Officer after just 5 months at @OpenAI. Stand by for updates.

@googrish congrats girish ying hang thariq and angel

“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof. today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data. open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise. with castform, model training is as simple as prompt engineering. @castformai bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns. see what you can build with castform👇

The difference between a billion and a trillion dollars is pretty much a trillion dollars.

@landay @drfeifei Congratulations, James!

Honored to lead this merged institute with a new vision and excited to have @drfeifei and John Hennessy advise me.

if we truly want models to do scientific discovery, we have to train them to not repeatedly attempt known ideas but be *exploratory*. they must search over vastly diverse, performant strategies under optimism and creative leaps of faith. @jubayer_hamid and i had a lot of fun chats when this was being conceptualized. glad it’s finally out. cc: @ifdita_hasan our RL pipelines should incorporate this in frontier agent training.

@heeney_luke math 175?

Everyone taking CS 153. Only 3 people in my Stanford functional analysis class today. Remember to eat your veggies.

@percyliang haha, congrats!

Academic titles are funny. After 14 years, I finally have the official title that people might have always assumed I had.

@typesfast Sec Lutnick is a funny guy

The Japanese won’t buy American cars except for the Yakuza who love Chevys.

we’re in the middle of a Cambrian explosion of new RL algorithms. this excellent (and extremely timely) work unifies many of their desirable properties. s/o to my childhood friend @fahimtajwar10 & team for setting a new direction in RLVR.

Link to our WSJ article: wsj.com/tech/a-big-dat…

Announcing our $35 Billion Project Concord 🇺🇸🇰🇷 Excited to unveil Project Concord with our first project: a $35 billion “Living Data Center” in South Korea — where AI designs, optimizes, and orchestrates the full stack, from silicon to systems to the entire site. Voltai is partnering with Stock Farm Road, an investor group founded by Brian Koo, the grandson of LG founder Koo In-hwoi, and Amin Badr-El-Din, founder of UAE Offsets Group, which spawned Mubadala, UAE’s sovereign wealth fund. Like Voltai, it also counts Alphabet Inc. (parent company of Google) Chairman John Hennessy as an advisor, alongside former Vice-President of the European Commission and Salesforce Board Member Neelie Kroes. Voltai and Stock Farm Road are the primary partners of Project Concord. This initiative aligns with South Korea's push to scale AI infrastructure. President Lee Jae Myung just announced plans to triple government AI spending in next year's budget. Project Concord marks the first time AI will serve as architect, manager, and operator of AI infrastructure that is fully vertically integrated, from silicon to rack to the whole facility. We're building the playbook for how AI infrastructure scales for the future of AI. News covered by The Wall Street Journal

@AshwinRamaswami gg

I passed!!

@laserboat999 clutter. picks up and drops off at your house albeit slightly $$. dm - happy to cover.