About Us

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-generated 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.

In October, we introduced Omnii-1, the world's most advanced multimodal genome language model, and a leap toward general biological intelligence. It unifies 28 biological modalities, spanning DNA, RNA, proteins, epigenomics, and more, in a single model trained on trillions of tokens of biological data In a world first, Omnii-1 designed the first AI-generated bacteriophages to be administered to a human patient with an antibiotic-resistant infection under compassionate use. We released this news alongside seven other project areas, showing the model’s breadth of applications.

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 Role

We are seeking research scientists and engineers in machine learning and artificial intelligence, with experience applying AI to biological problems or an interest in doing so. You will help lead the design, evaluation, and deployment of our multimodal genome language models, with a particular focus on our biosecurity and biodefense efforts.

In this role, you will collaborate with our multidisciplinary team (e.g., computational biologists, AI scientists, and systems architecture engineers) to translate advances in large-scale machine learning into capabilities for modeling biological systems. You will extend and adapt large model backbones—such as sequence and multimodal foundation models—to enable tasks across genomics, protein biology, and cellular systems. This includes designing post-training pipelines, domain adaptation strategies, and evaluation frameworks that enable state-of-the-art ML frameworks to reason over biological data.

What You'll Do

Build and adapt frontier AI models through fine-tuning and post-training.

Design and run experiments applying large models to problems relevant to biosecurity and biodefense.

Bridge to real-world biosecurity systems. Translate technical capabilities into deployable systems such as early warning platforms, sequence screening tools, attribution technologies, and rapid detection and response pipelines. Work with government and partners to align systems with real-world workflows.

Analyze model behavior and run ablations to understand model reasoning and failure modes in biological contexts.

Explore modern mechanistic interpretability pipelines and methods for sequence characterization and model understanding.

Prototype new approaches for prediction and design using foundation models.

Operate with high ownership in a fast-moving, ambiguous environment.

What We're Looking For

Strong technical background in deep learning for large-scale classification, retrieval, or generative models.

Ability to design evaluation tasks and benchmarks that measure model capability beyond simple accuracy metrics, with a critical eye toward aligning computational outputs with actionable downstream applications.

Strong Python and ML tooling experience (PyTorch or JAX, experiment management, distributed training)

Experience applying ML models to biological data and challenging prediction tasks.

Familiarity with molecular biology and biological data modalities, particularly genomics, gene regulation, protein biology, or cellular systems, or an interest in applying your AI/ML skills in this space.

Ability to reason concretely about what biological AI systems can and cannot do, how those capabilities translate into real-world risks, and how to measure and mitigate those risks.