Normal Computing 🧠🌡️

432 posts

Normal Computing 🧠🌡️ banner
Normal Computing 🧠🌡️

Normal Computing 🧠🌡️

@NormalComputing

We build AI systems that natively reason, so they can partner with us on our most important problems. Join us https://t.co/G90CO1W456

New York, NY Se unió Haziran 2022
124 Siguiendo4.4K Seguidores
Normal Computing 🧠🌡️
Normal Computing 🧠🌡️@NormalComputing·
Our Head of AI @thomasahle ran agents autonomously for 43 days and built a full verification stack: simulation, UVM, formal, mutation testing. 580K lines. The full writeup covers what worked, what didn't, and why formalized problems with objective correctness signals may be uniquely well-suited to agentic AI: normalcomputing.com/blog/building-…
English
8
17
220
2M
Normal Computing 🧠🌡️ retuiteado
Casper Hansen
Casper Hansen@casper_hansen_·
thrilled to share that i joined @NormalComputing as an ai engineer in their copenhagen office the company is not only building thermodynamic hardware with up to 1000x energy efficiency, but also an EDA product for building new hardware this is a future that i'm excited for!
Casper Hansen tweet media
English
11
3
75
5.9K
Normal Computing 🧠🌡️ retuiteado
Thomas Ahle
Thomas Ahle@thomasahle·
my results on AI for autonomous progress: we ran codex+claude for 43 days straight to build a System Verilog compiler/simulator normalcomputing.com/blog/building-…
English
15
43
301
51.3K
Normal Computing 🧠🌡️ retuiteado
Sam Duffield
Sam Duffield@Sam_Duffield·
New open source: cuthbert 🐛 State space models with all the hotness: (temporally) parallelisable, JAX, Kalman, SMC
Sam Duffield tweet media
English
1
13
63
9.3K
Normal Computing 🧠🌡️
Normal Computing 🧠🌡️@NormalComputing·
We’re sharing Thermodynamic Natural Gradient Descent, now published in Nature Unconventional Computing. Second-order optimization methods like natural gradient descent promise faster, more stable training, but are rarely used at scale because computing the natural gradient is prohibitively expensive on conventional hardware. This work shows the limitation is physical, not algorithmic, using a thermodynamic computing element to compute the natural gradient via physical dynamics. Paper: nature.com/articles/s4433…
English
3
12
99
7.1K
Normal Computing 🧠🌡️
Normal Computing 🧠🌡️@NormalComputing·
Demand for more efficient AI computing is driving the silicon complexity crisis. Normal EDA was built for this reality. As teams build energy-efficient chips for target workloads, design complexity and iteration cycles keep growing. Modern EDA needs a unified view across the flow to optimize across abstraction layers and enable workload-driven architectures. Perhaps even unconventional ones. See how teams use its Verification Suite in production to generate verification artifacts, close coverage faster, reduce manual work, and head into signoff with confidence.
English
2
0
10
72.9K
Normal Computing 🧠🌡️
Normal Computing 🧠🌡️@NormalComputing·
Proud to be included in @ARIA_research 2025 lookback as part of the Scaling Compute programme. This year, we taped out the world’s first thermodynamic computing chip, realizing this computing approach in silicon for the first time. ARIA’s support made it possible to pursue a physics-based alternative to conventional architectures. Grateful to the ARIA team for backing long-horizon, first-principles engineering. aria.org.uk/insights/a-loo…
English
2
1
4
28.3K
Normal Computing 🧠🌡️ retuiteado
Max Ottignon
Max Ottignon@MaxOttignon·
@amazon @studiokoto @NormalComputing by Company Policy. All AI brands look the same? Not this one. This is B2B with swagger. The name. The brash red. The sophisticated design system. The just-the-right-side-of-arrogant key line: “when the future of AI arrives, it will be Normal.”
Max Ottignon tweet media
English
1
1
3
609
Normal Computing 🧠🌡️ retuiteado
Alex Tong
Alex Tong@AlexanderTong7·
@gavincrooks from @NormalComputing taking the #FPI stage at #Neurips2025. ⚛️💻 Why spend massive energy simulating probabilistic sampling on deterministic GPUs when we can build hardware that is natively probabilistic?
Alex Tong tweet media
English
1
5
29
2.5K
Normal Computing 🧠🌡️
Normal Computing 🧠🌡️@NormalComputing·
We are pleased to announce that our CEO, @FarisSbahi, will be presenting on behalf of @NormalComputing at the @ARIA_research stand at #NeurIPS2025. He will be speaking today at 19:00 PM at booth 1343. Faris will share our latest R&D work with ARIA, including how our CN101 thermodynamic computing chip, recently taped-out, rewrites the energy-compute trade-off for AI. If you are attending NeurIPS, we encourage you to stop by. We look forward to meeting researchers, collaborators, and anyone interested in the future of AI.
English
1
1
8
798
Normal Computing 🧠🌡️ retuiteado
Gavin Crooks
Gavin Crooks@gavincrooks·
Inbound to Neurips @ San Deigo. @FarisSbahi will be speaking about @NormalComputing at the ARIA/UK DBT stand at the expo, Tuesday at 19:00. (We're hiring!) I'm speaking on Sunday at 13:15, Frontiers in Probabilistic Inference workshop. Let's do this thing.
English
0
4
28
4.3K
Normal Computing 🧠🌡️ retuiteado
Erik Meijer
Erik Meijer@headinthebox·
AI for hardware goes *much* deeper than AI for software.
Thomas Ahle@thomasahle

OK super narrow hiring post for a high-prio role (actually, there may not be anyone in the world who fits this) but if you: - have some familiarity with both RL+agents, and have gone deep on at least one - have experience with hardware engineering, device verification in particular - are ready to dive into the insanely deep hole that is AI for hardware - are already based in CPH or NY, or eager to move there asap - can start a new FT role within 30 days - are interested in going deep on application-specific agents - lean extroverted, and are comfortable with public-facing comms - are keen to join a fast-moving + fast-growing startup DM me with “APPLICATION” as the first token, followed by evidence for the above do NOT use that token for any other role inquiries (your DM will be deleted); apply through links, or DM others on the team. same goes for replies. this is far from the only role open, but it’s the one i personally have the highest vested interest in finding a killer fit for it + will be prioritizing it when reviewing DM apps it’s a pretty specific list, and i’m looking for high alignment on ~all of the above, but if your life journey has somehow led you to this specific set of skills, there likely isn't anywhere you can get more use of them signal-boosts appreciated. it’s a very high-leverage position + will be compensated accordingly :)

English
0
4
16
6.4K
Normal Computing 🧠🌡️ retuiteado
Thomas Ahle
Thomas Ahle@thomasahle·
Everyone knows RLHF and RLVR, but do you know the 17 other RLXX methods published in papers and blog posts? Here's a condensed list:
English
3
4
22
2.8K
Normal Computing 🧠🌡️ retuiteado
Anastasis Germanidis
Anastasis Germanidis@agermanidis·
Diffusion for everything! We share a recipe to start from a pretrained autoregressive VLM and, with very little training compute and some nice annealing tricks, turn it into a SOTA diffusion VLM. Research in diffusion for language is progressing very quickly and in my mind, provides as promising of a path of unifying modalities as the 'omni' autoregressive models. Amazing work led by @mariannearr @ServerProcessor over the summer.
Runway@runwayml

Today we're sharing our first research work exploring diffusion for language models: Autoregressive-to-Diffusion Vision Language Models We develop a state-of-the-art diffusion vision language model, Autoregressive-to-Diffusion (A2D), by adapting an existing autoregressive vision language model for parallel diffusion decoding. Our approach makes it easy to unlock the speed-quality trade-off of diffusion language models without training from scratch, by leveraging existing pre-trained autoregressive models.

English
6
28
283
39.8K
Normal Computing 🧠🌡️
Normal Computing 🧠🌡️@NormalComputing·
We’re pleased to announce that @NormalComputing, alongside the John Templeton Foundation, the State of Maryland’s Capital of Quantum Initiative, University of Maryland partners, and @Fidelity Investments, have pledged hands-on support to the Maryland Quantum-Thermodynamics Hub (@JointQuICS), which has just secured over $5M in new funding to expand its pioneering research into thermodynamics, energy flow, and the physics of time. Since its launch in 2022, the hub has: -Published 60+ papers and supported multiple PhD theses, - Brought together researchers across UMD, UMBC, USC, University of Arizona, Los Alamos, University of Rochester, and University College Dublin, - Explored deep thermodynamic questions—from the arrow of time to how quantum systems thermalize, exchange energy and information with their environments, and ultimately impact error correction in quantum computing. At Normal Computing, our mission is to co-design algorithms and hardware aligned with the physics of noise, energy, and fluctuations. Supporting the hub is part of that vision to advance fundamental thermodynamics research that explains how systems exchange energy and information, and applying those insights to build the next generation of computing.
English
1
1
17
1.9K

Descubrir

@thomasahle @zierhjmensch @adnarim066 @wellingmax @Sam_Duffield @MaxAifer @ColesThermoAI @ARIA_research