Member of Technical Staff, Protein Design

Radical Numerics•San Francisco, CA

About The Position

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.

Requirements

Strong experience developing machine-learning models for protein structure prediction, structural biology, geometric deep learning, or a closely related area.
Experience training or fine-tuning protein language models, structure models, diffusion models, or other large scientific machine-learning systems.
Deep understanding of modern protein structure-prediction and design methods, loss objectives, and architectures.
Familiarity with geometric neural networks, equivariant architectures, pairwise representations, and generative modeling of molecular structure.
Strong knowledge of protein structure, including secondary and tertiary structure, protein domains, complexes, conformational flexibility, and evolutionary constraints.
Experience building and validating biological datasets and controlling for data leakage, homology, and benchmark contamination.
Familiarity with commonly used protein structure metrics and evaluation practices.
Fluency in Python and a modern deep-learning framework such as PyTorch or JAX.
Experience with distributed training, accelerators, large datasets, and reproducible experimentation.
Strong experimental judgment and the ability to distinguish genuine scientific progress from benchmark artifacts.
Ability to independently move between research, implementation, experimentation, and scientific analysis.
Clear communication skills and an ability to collaborate across machine learning, computational biology, and engineering.
Expertise in machine learning, computational biology, structural biology, biophysics, computer science, or a related field, or an equivalent record of research and engineering impact.

Nice To Haves

Contributions to protein structure-prediction systems, protein foundation models, geometric generative models, or widely used structural biology software.
Familiarity with multiple sequence alignments, templates, coevolutionary methods, inverse folding, molecular simulation, or energy-based modeling.
Experience modeling other macromolecules, small molecule systems, and molecular interactions.
Understanding of over major protein structure datasets, benchmarks, or community evaluation efforts.
Experience modeling protein complexes or alternative conformational states.
Experience with SE(3)- or E(3)-equivariant architectures, diffusion models, flow matching, or generative modeling of molecular coordinates.
Experience evaluating model confidence, uncertainty, and calibration.
Familiarity with experimental methods for determining protein structure.
A record of publications, open-source contributions, or production systems demonstrating impact in generative modeling, molecular design, or scientific machine learning.

Responsibilities

Develop and improve machine-learning models for protein structure prediction, design, and related structural biology tasks.
Train and fine-tune protein language models, geometric neural networks, diffusion models, and other modern scientific machine-learning architectures.
Explore new architectures and learning objectives for modeling protein sequence and structure.
Build reliable data pipelines and evaluation systems for structural modeling.
Design rigorous benchmarks that measure generalization and minimize data leakage or memorization.
Evaluate models using established structural accuracy, confidence, and physical-validity metrics.
Analyze model performance across diverse proteins, structural classes, and biological contexts.
Run ablation studies and controlled experiments to understand the impact of model architecture, data, scale, and training methodology.
Improve the efficiency and reliability of model training and inference on large-scale compute systems.
Collaborate with scientists and engineers to translate research advances into robust modeling capabilities.