Radical Numerics is an AI research lab building general biological intelligence. Our mission is to master the code of life, and our purpose is to reduce human suffering.
Our team created Evo, and started the field of generative genomics. Our work was featured on the cover of Science, and presented by our CEO on the main stage of TED2025. Evo was used to create the first AI gene therapy tool CRISPR-Cas9, and the first AI whole genome from scratch. Evo 2, featured in Nature, is the largest fully open source AI project across any domain.
Radical Numerics is bringing the rigor of distributed systems, model architecture, and numerics research to the challenges of biology. We’ve redesigned the foundation model training stack to turn the world’s raw scientific data (e.g. biological sequences, experiments, and physical processes), into intelligible, generative models that can expand and accelerate what humanity can understand, design, and cure.
The same generative breakthroughs that enable life-saving cures also lowers the barrier to creating engineered threats and AI-generated bioweapons. We believe these forces are inseparable. Radical Numerics was founded to develop both the power to design and the responsibility to defend.
About the RoleYou build what our partners use. Our models produce variant scores; you build everything between those scores and the scientist who needs to act on them.
The science team owns the model itself. You own the pipelines that run it on partner data, the interfaces people use to work with the results, and the evaluations that demonstrate how the scores hold up. Our inference infrastructure and core API are out of scope for this role.
You'll be paired with a product lead who owns a customer lane. You'll present your own work directly in partner meetings, which means fielding technical objections yourself. Expect to ship on a weekly cadence.
Develop the application surface for your area of ownership, including interactive workbenches, variant scoring and ranking pipelines, and other tools that make model output usable by scientists unfamiliar with our API.
Run our models on partner data and present results with enough interpretability for a computational biologist to trust the output.
Build the evaluation infrastructure that makes our claims verifiable, including held-out benchmarks, blind retrospectives on solved cases, and head-to-head comparisons against tools partners currently use.
Present your own work directly in partner meetings, field technical objections in real time, and return with fixes.
Ship on a weekly cadence. Most features begin as prototypes developed live in partner conversations.
You build across the stack. Python for data and model work, TypeScript, React or similar for the interfaces.
You have worked with real genomic data. VCFs, reference builds, variant annotation, and the specific ways this data is messy in practice.
Comfortable engaging a partner's computational biology team as a technical peer, including in skeptical or challenging technical discussions.
Comfortable operating without a finished spec, and able to take a rough direction and a partner conversation and turn them into a working system within a week.
Experience designing model evaluations, and an opinion about what makes a benchmark honest.
You've handled sensitive human data under strict access controls and understand the limits that comes with.
A forward deployed engineering, solutions engineering, or founding engineer background at an AI or infrastructure company.
Radical Numerics is committed to equal employment opportunity and does not discriminate in any employment opportunities or practices based on an individual's race, color, creed, gender (including gender identity and gender expression), religion (all aspects of religious beliefs, observance or practice, including religious dress or grooming practices), marital status, registered domestic partner status, age, national origin or ancestry (including language use restrictions and possession of a driver’s license issued under California Vehicle Code section 12801.9), natural hair, physical or mental disability, political affiliation, medical condition (including cancer or a record or history of cancer, and genetic characteristics), sex (including pregnancy, childbirth, breastfeeding or related medical condition), genetic information, sexual orientation, military and veteran status or any other consideration made unlawful by federal, state, or local laws. It also prohibits unlawful discrimination based on the perception that anyone has any of those characteristics, or is associated with a person who has or is perceived as having any of those characteristics.
Radical Numerics participates in E-Verify and will provide the federal government with your Form I-9 information to confirm that you are authorized to work in the U.S.
Skills Required
- Proficient in Python for data and model work
- Proficient in TypeScript and React for building interfaces
- Experience working with real genomic data (VCFs, reference builds, variant annotation)
- Able to engage computational biology teams as a technical peer and field technical objections
- Able to operate without a finished spec and deliver working systems within a week
- Present work directly in partner meetings and iterate based on feedback
- Experience designing model evaluations and benchmarks
- Experience handling sensitive human data under strict access controls
- Background in forward deployed engineering, solutions engineering, or founding engineer at AI/infrastructure company
What We Do
The architectures that power modern AI were designed for language, not for science. They were never built to understand DNA, interpret experiments, or learn the rules that govern living systems. Closing this gap represents one of the defining opportunities of our time. Our team has pioneered advances across frontier AI, including the first models trained with one-million-token context windows, and now scaling toward one billion. We focus on innovations on systems and architecture, because building AI for science demands a fundamentally different foundation. As a first application of this technology engine, we are advancing AI for life, the most complex and consequential domain of all. We are the AI team behind Evo and Evo 2, the generative genomics models used to create real gene-editing tools and the first whole genomes designed entirely by AI. That work demonstrated that AI can create biology, not just analyze it. We are now building multimodal models trained directly on the fabric of biology, enabling faster discovery, deeper understanding, and entirely new capabilities. As we push the frontier of general biological intelligence, we will build systems hand-in-hand to ensure global biological resilience: rapid detection, rapid response, and rapid countermeasures against emerging threats, both natural and synthetic. Our mission is ambitious: to reimagine what AI can do for biology, and to build that future. Our advisors include Eric Horvitz, CSO of Microsoft, Chris Ré of Stanford, George Church of Harvard, and Andrew Weber, former Assistant Secretary of Defense for Nuclear, Chemical and Biological Defense Programs.
Why Work With Us
We bring the rigor of distributed systems, model architecture, and numerics research to the hard problems of scaling learning on biological data. If this resonates, we'd love to hear from you.
Gallery
