Co-Founder @CallosumAI building Intelligent Systems through co-design of novel hardware & algorithms, at least loosely inspired by nature. Also @Cambridge_Uni.
With our $100M seed round, we announce our first wave of global compute partnerships, flagship among them @CerebrasSystems.
Partnerships spanning silicon, infrastructure and geography. The path to intelligence that is orders-of-magnitude better, cheaper, and more sovereign.
It can but the way I think about it is that a CPU would, due to its generality, assume every possible graph structure as equally likely at each stage, both across the graph topology and the operations on graph; following standard specificity-generality tradeoffs you end up wasting energy at each traversal step on not using operation-topology-structure. How much of a waste that is depends a lot on the application! However graphs in the real world tend to be quite structured, so no reason to believe there isn't sufficient structure in all of the computational graphs here too.
my heterogeneous compute thesis in one gif, and what i'm investing against. lots of chips each good at one bit of the job, plus something routing the work. @CallosumAI does the routing, @TendrilsCompute the CPU for agents, @fractile_ai fast decode. graph chip still missing
Sam Altman created the most ruthless model in the world
we built the biggest mafia-like game ever
> 250.000 humans playing with/against AI
we found:
@OpenAI → maniac LIARS
@spacexai → by far the most LOYAL
@AnthropicAI → the biggest SNITCH
we also added a flybrain ;)
Our CEO and co-founder @DanAkarca was live this morning on @BloombergTV, discussing the evolving chip landscape and the future of AI infrastructure in the context of recent NVIDIA news.
At Callosum, we’re building heterogeneous compute systems that work with any AI chip, across @nvidia and our wider network of hardware partners. Our unique approach co-evolves chips and intelligence together, unlocking new levels of capability and efficiency.
Every processor basically works the same way. It takes a list of instructions and goes through them in order, one after another. It's a design from the 1940s.
What if neither CPUs nor GPUs are the right chips for agents?
Why we invested:
https://t.co/ez9owD96m7
Tendrils has raised $30M led by @benchmark to rebuild general-purpose computing. @jaltma is joining our board.
At Tendrils, we believe that the future of general-purpose computing lies beyond legacy architectures for legacy software.
Tendrils has raised $30M led by @benchmark to rebuild general-purpose computing. @jaltma is joining our board.
At Tendrils, we believe that the future of general-purpose computing lies beyond legacy architectures for legacy software.
Couldn't be happier to announce my new paper on defining compositional structure — and generating it! I see this project as the culmination of my PhD, as it unifies all my interests: compositionality, compression, structure, and open-endedness. More below! https://t.co/G2T6sOQDGc
We have been named in the Budget’s prospectus 2026, as one of the most strategically important companies backed by the UK government.
We’re building heterogeneous compute: bringing different AI chips and models together into complete systems.
Thank you to the @UKSovereignAI and @ARIA_research.
We are just getting started.
Our CTO @achterbrain joined @nvidia on stage at the start-up day of GTC Berlin, alongside Clara Cornu, @talfanevans, @nouralhassan and Matteo Sarcinella.
He discussed how Callosum runs heterogeneous workloads efficiently across every generation and type of NVIDIA hardware, and what it takes to scale new technology companies fast in Europe.
Thanks for having us @nvidia
We’ve already backed some of the most ambitious AI companies being built in Britain.
From AI infrastructure and safety to science and new technologies, our portfolio spans some of the most exciting areas of AI, all with ambitious founders building from the UK.
Let’s hear from a few of the companies we’ve backed so far.
@CallosumAI | Augur | @Basecamp_Res
@WoodingtonBen and the @coherenceneuro team are doing absolutely amazing work! Could not be more excited to see them raising their Series A; more than well deserved!
We’ve raised a $45m Series A, co-led by
@foundersfund and The @OpenAI Startup Fund, to build cancer therapy that learns. Technology is rapidly changing medicine, but data loops and rapid adjustments are still largely missing. When you start cancer treatment, knowing whether what you’re doing is working is too slow. Clearly there is a better way. Build systems to watch cancer in real time, then feed that data into a system that itself can treat the disease. This seeds a future where, unlike drugs, we develop medicine that continuously learns to improve itself. We've been building neurotechnology and BCI platforms to do just that.
@coherenceneuro
We’ve raised a $45m Series A, co-led by
@foundersfund and The @OpenAI Startup Fund, to build cancer therapy that learns. Technology is rapidly changing medicine, but data loops and rapid adjustments are still largely missing. When you start cancer treatment, knowing whether what you’re doing is working is too slow. Clearly there is a better way. Build systems to watch cancer in real time, then feed that data into a system that itself can treat the disease. This seeds a future where, unlike drugs, we develop medicine that continuously learns to improve itself. We've been building neurotechnology and BCI platforms to do just that.
@coherenceneuro