AI, numerics and systems. Co-founder & Chief AI Scientist @RadicalNumerics.
Joined August 2018
- Tweets458
- Following257
- Followers3.5K
- Likes2.1K
Michael Poli retweeted
Language models can learn new tasks from just a few examples in their prompt, with no retraining. This is called in-context learning (ICL). Now genome language models can do it too.
With a handful of examples, Evo 2 ICL generates proteins, de novo antibodies, CRISPR-Cas9 loci, mitochondrial genomes and entire bacteriophages.
We can even use the synthetic sequences iteratively to construct a chain-of-thought (CoT) and achieve harder design goals.
New work with Yukun A. Hao, @BrianHie and team 🧵
Michael Poli retweeted
1/ With Omnii Diagnostics, we asked whether a multimodal genome language model could help find cancers whose signals in blood are too faint for current classifiers.
We adapted Omnii-1 to a biomarker never seen in pretraining. In a test cohort, Omnii's features improved early-stage cancer detection
Michael Poli retweeted
Super excited about this release. One of the key advantages Omnii-1 has is its flexibility in accommodating new data. We explore some vignettes in our Omnii Genome blog post, and it’s also been a lot of fun pointing Omnii-1 at totally new tasks such as cancer diagnostics. (1/4)
1/10 Introducing Omnii Genome, one of eight Advances built on Omnii-1.
A single generalist model now outperforms leading specialized models (AlphaGenome) in linking genetic changes to gene activity and mapping regulatory interactions.
These capabilities help make sense of the 98% of the human genome that doesn’t code for proteins
How Omnii-1 works:
A deeper dive on Omnii Genome. On a personal level, it's been extremely exciting to plan around the new possibilities enabled by a generative paradigm here.
1/10 Introducing Omnii Genome, one of eight Advances built on Omnii-1.
A single generalist model now outperforms leading specialized models (AlphaGenome) in linking genetic changes to gene activity and mapping regulatory interactions.
These capabilities help make sense of the 98% of the human genome that doesn’t code for proteins
How Omnii-1 works:
Today, we are launching Omnii-1, our next-generation multimodal genome language model.
Omnii-1 draws on a long line of research breakthroughs in genome language modeling, spanning model architectures, multimodal pretraining, as well as new mid-training & post-training methods tailored to biological data.
The launch includes a staggering amount of technical material: 8 full-length technical blogs (and we are just getting started).
Proud of the team for this milestone!
Michael Poli retweeted
1/8 Does text-to-image generation need language pretraining?
Today, we share Halcyon, a family of early-fusion, base models trained from scratch. Halcyon outperforms strong late-fusion baselines with up to 275× less pretraining compute. Models and recipes are open.
Halcyon is part of a research program at Radical Numerics on the science of multimodal pretraining and scaling.
Michael Poli retweeted
🚀 Introducing Next-Concept Modeling (NCM): our new paradigm for fast, parallel, coarse-to-fine text generation.
Compared to strong diffusion language models (DLMs) at similar parameter size, NCM achieves:
⚡ Up to 31× faster generation on downstream tasks like summarization and machine translation with comparable or better quality.
🏎️ 10–21× faster unconditional generation across TinyStories, LM1B, and OpenWebText, with comparable or better generative perplexity and using 3–5× fewer training tokens.
🧩 Better performance on reasoning tasks like Sudoku, outperforming both AR and diffusion baselines with fewer sampling steps.
With @ishanskhare, @IddahMlauzi, @ekellbuch, @simran_s_arora, @Azaliamirh, and @HazyResearch.
More details in the thread 🧵👇
Michael Poli retweeted
1/5 What does it take to scale pretraining to 20+ data modalities at once? Improved data mixing methods are one important ingredient.
Today, we share how we scale data mixing to thousands of sources for DNA and natural language pretraining of Omnii, our multimodal genome model.
Michael Poli retweeted
Genomes are full of dark matter of unknown functions.
We present Minerva, a new method for discovery guided by genome language models. Minerva reveals the interactions hidden in non-coding DNA, pointing to hundreds of new putative RNAs and repetitive elements per bacterial genome. Minerva allows us to find and study elements invisible to traditional methods at orders of magnitude greater scale than before.
Just a taste of what Omnii can do.
1/5 We pretrained Omnii, a multimodal language model on DNA. Then, we asked it to design a cancer vaccine
In new research, we show how Omnii learned to turn a tumor DNA sequence into a personalized mRNA vaccine design.
Read more here: radicalnumerics.ai/blog/omni…
Michael Poli retweeted
Radical Numerics has 16 roles open, from MTS pre/post training, infra & systems, mech interp, to product leads in scientific discovery.
Our team is growing rapidly in SF, and so are the models. We’re pretraining the largest biological AI models ever. Science is the next frontier in AI.
Roles:
radicalnumerics.ai/join-us
- MTS Pretraining
- MTS Post-training
- MTS Infra & systems
- MTS Model interpretability
- MTS ML engineer
- MTS Inference
- MTS AI supercomputing
- Product lead - Discovery
- Product lead - Diagnostics
- Computation biology
- Protein design
- MTS Biosecurity
- Head of Data Acquisition
- MTS AI bio
- Applications engineer
- Marketing associate
Showcasing a critical application of our new frontier genome language model, Omnii.
Omnii can strengthen biodefense pipelines by covering blind spots of existing search-based methods.
1/ Leaders of OpenAI, Anthropic, & GDM signed a letter urging Congress to mandate DNA synthesis screening.
But chatbots & agents can't read DNA.
So we built Omnii to defend against natural & AI-designed pathogens.
Blog: radicalnumerics.ai/blog/omni…
Letter: screendna.org
Michael Poli retweeted
Some of the best biologists that I’ve ever worked with have joined this team at Radical Numerics, I have a lot of confidence in them and their ability to get genomic language modeling right. Check them out if you want to work at the frontier of genome design and bio defense!
Together with my co-founders Michael @MichaelPoli6, Stefano @Massastrello and Armin @athmsx, I am excited to announce @RadicalNumerics is emerging from stealth with a $50M seed round to build general biological intelligence.
We’re also sharing an early preview of our new model Omnii, the most powerful genome language model to date.
Omnii preview link:
radicalnumerics.ai/blog/radi…
At Radical Numerics, our mission is to master the code of life, and to drive the frontier of biological AI for both design and defense.
This is our dual mandate, which comes from something our own team helped make possible.
Our founding team trained Evo and Evo 2, the largest biological AI models (40B params) trained on DNA sequences. Trillions of tokens across all of life, from microbes to mammals. It’s fully open source, and created the field now known as generative genomics.
Last year, scientists used Evo to generate the world’s first complete genome from scratch using AI. Turns out it was a bacteriophage—a type of virus. It functioned in the real world, and in this case it was harmless. But for us, it was a clear turning point.
It showed that AI is no longer just analyzing biology. It is on the cusp of generating functional lifeforms. Eventually, AI will have the power to design and control life itself.
That should make all of us incredibly excited, and incredibly uneasy. (Anyone can design DNA with a new function, and have it synthesized and delivered, like something from Amazon Prime).
The same technology that will help us cure cancer is the very technology that might create the next global pandemic, or worse, allow the creation of bioweapons that can wipe out populations.
We believe these forces are inseparable. If you work on the frontier of biology, you have to build technology to safeguard it from its misuse. Existing biosecurity tools are sorely losing the arms race, relying on outdated “have I seen this exact thing before?” style algorithms.
We founded Radical Numerics to turn the tide.
And we can’t do that by training on textbooks and natural language. We must understand the language of biology from the raw physical data itself, to reason across every molecule and modality, from DNA to proteins.
The next frontier for AI goes far beyond chatbots or video generators to models that can understand and engineer life.
Today, we’re previewing Omnii, which is already far surpassing Evo 2, and will continue improving as we scale and add new modalities (training now).
1. For human health, Omnii can read and write whole genomes (more on writing later). It’s state of the art (SOTA) on detecting causal variants for disease, and can rank Alzheimer's mutations zero-shot. We’re partnering with a diagnostics company to use Omnii for early cancer detection (pancreatic and multi-cancer).
2. For defense, Omnii is SOTA at detecting AI-generated pathogens. We benchmarked existing detection tools, and they simply can’t detect the AI-generated ones (“deepfake viruses”). We’re partnering with a US national lab to pilot Omnii for detecting the next pandemic, both natural and AI-generated.
We have a data center full of Blackwells in construction now to build the most powerful biological AI models ever. This mission takes a new kind of AI lab that can actually scale on physical, biological data: new alignment research (mid/post training), scaling long context, building out mech interp teams to dissect what these models learn, new architectures and systems designs, all from the ground up.
Our team is made up of AI researchers and scientists from top labs and institutions (e.g. Stanford, MIT, Google DeepMind), but more importantly, we all share the belief that this is the most important challenge of our lifetime. If you feel similarly, we are hiring. We aim to bring the brightest minds in AI and science together to save lives.
Thanks to our partners on this journey, led by Emergence Capital @emergencecap, with Obvious Ventures @obviousvc, Triatomic @TriatomicCap
, and Patrick Collison @patrickc. Our advisors include Eric Horvitz @erichorvitz, CSO of Microsoft, Chris Re @HazyResearch of Stanford, George Church @geochurch of Harvard, and Andrew Weber @AndyWeberNCB, former Assistant Secretary of Defense for Nuclear, Chemical and Biological Defense Programs.
Fortune article: fortune.com/2026/06/15/exclu…
Jobs: radicalnumerics.ai/join-us
Excited to continue the journey with Eric, Stefano, Armin, the team at @RadicalNumerics, and to have the chance to preview our next-generation model, Omnii.
True progress in biological AI demands a culture where the same conversation routinely covers frontier pretraining, systems, and data mixing one moment and complex biology the next. If that sounds like a blast, consider joining us.
Together with my co-founders Michael @MichaelPoli6, Stefano @Massastrello and Armin @athmsx, I am excited to announce @RadicalNumerics is emerging from stealth with a $50M seed round to build general biological intelligence.
We’re also sharing an early preview of our new model Omnii, the most powerful genome language model to date.
Omnii preview link:
radicalnumerics.ai/blog/radi…
At Radical Numerics, our mission is to master the code of life, and to drive the frontier of biological AI for both design and defense.
This is our dual mandate, which comes from something our own team helped make possible.
Our founding team trained Evo and Evo 2, the largest biological AI models (40B params) trained on DNA sequences. Trillions of tokens across all of life, from microbes to mammals. It’s fully open source, and created the field now known as generative genomics.
Last year, scientists used Evo to generate the world’s first complete genome from scratch using AI. Turns out it was a bacteriophage—a type of virus. It functioned in the real world, and in this case it was harmless. But for us, it was a clear turning point.
It showed that AI is no longer just analyzing biology. It is on the cusp of generating functional lifeforms. Eventually, AI will have the power to design and control life itself.
That should make all of us incredibly excited, and incredibly uneasy. (Anyone can design DNA with a new function, and have it synthesized and delivered, like something from Amazon Prime).
The same technology that will help us cure cancer is the very technology that might create the next global pandemic, or worse, allow the creation of bioweapons that can wipe out populations.
We believe these forces are inseparable. If you work on the frontier of biology, you have to build technology to safeguard it from its misuse. Existing biosecurity tools are sorely losing the arms race, relying on outdated “have I seen this exact thing before?” style algorithms.
We founded Radical Numerics to turn the tide.
And we can’t do that by training on textbooks and natural language. We must understand the language of biology from the raw physical data itself, to reason across every molecule and modality, from DNA to proteins.
The next frontier for AI goes far beyond chatbots or video generators to models that can understand and engineer life.
Today, we’re previewing Omnii, which is already far surpassing Evo 2, and will continue improving as we scale and add new modalities (training now).
1. For human health, Omnii can read and write whole genomes (more on writing later). It’s state of the art (SOTA) on detecting causal variants for disease, and can rank Alzheimer's mutations zero-shot. We’re partnering with a diagnostics company to use Omnii for early cancer detection (pancreatic and multi-cancer).
2. For defense, Omnii is SOTA at detecting AI-generated pathogens. We benchmarked existing detection tools, and they simply can’t detect the AI-generated ones (“deepfake viruses”). We’re partnering with a US national lab to pilot Omnii for detecting the next pandemic, both natural and AI-generated.
We have a data center full of Blackwells in construction now to build the most powerful biological AI models ever. This mission takes a new kind of AI lab that can actually scale on physical, biological data: new alignment research (mid/post training), scaling long context, building out mech interp teams to dissect what these models learn, new architectures and systems designs, all from the ground up.
Our team is made up of AI researchers and scientists from top labs and institutions (e.g. Stanford, MIT, Google DeepMind), but more importantly, we all share the belief that this is the most important challenge of our lifetime. If you feel similarly, we are hiring. We aim to bring the brightest minds in AI and science together to save lives.
Thanks to our partners on this journey, led by Emergence Capital @emergencecap, with Obvious Ventures @obviousvc, Triatomic @TriatomicCap
, and Patrick Collison @patrickc. Our advisors include Eric Horvitz @erichorvitz, CSO of Microsoft, Chris Re @HazyResearch of Stanford, George Church @geochurch of Harvard, and Andrew Weber @AndyWeberNCB, former Assistant Secretary of Defense for Nuclear, Chemical and Biological Defense Programs.
Fortune article: fortune.com/2026/06/15/exclu…
Jobs: radicalnumerics.ai/join-us
The next ImageNet for visual generative model pretraining research. Very important work, led by @keshigeyan and @KyleSargentAI!
1/ Introducing GPIC: a Giant Permissive Image Corpus and benchmark for visual generation!
🚀100M VLM-captioned image-text pairs for training
📊1M image-text pairs for benchmarking
🖼️~28 trillion pixels
🤗Centrally Hosted
✅Fully permissive for research + commercial use
Dataset, benchmark and models🧵👇
Co-led with @KyleSargentAI
We're growing rapidly at @RadicalNumerics and scaling our core teams. Join us in building the next generation of scientific world models.
We're hiring across a few roles, each with significant ownership and cross-functional scope:
- Member of Technical Staff, Post-Training
- Member of Technical Staff, Infrastructure and Training Systems
- Member of Technical Staff, Pretraining Science
- Member of Technical Staff, AI Bio
- Member of Technical Staff, Biosecurity
Our technology brings together numerics, systems engineering, and architecture design to tackle large-scale pretraining on scientific data. Our blogs (see below) give a flavor of the work.
We believe that advancing capabilities must go hand-in-hand with advancing safety and biosecurity. The same systems that design biology must also help defend against it.
Ping me or others in the team if you'd like to learn more.
job-boards.greenhouse.io/rad…
