EPM Scientific
Machine Learning Researcher - Generative Modeling for Drug Discovery
A talent-dense AI4Biology team is applying frontier-scale AI to model biological systems and accelerate therapeutic development. They are building novel architectures and generative frameworks to reason about molecular behavior, predict drug toxicity and potency, and simulate complex biological interactions.
This role is suited for a thoughtful and curious scientist with deep engineering rigor, focused on designing and training models that understand and generate molecular structure, dynamics, and function. It's a zero-to-one opportunity to shape the technical foundation of a company working at the intersection of AI and human health.
Apply now, read the job details by scrolling down Double check you have the necessary skills before sending an application.
Focus Areas: Designing generative models (e.g. diffusion, flow matching, graph-based) for molecular structure and dynamics Building predictive systems for ADMET properties, drug-likeness, and off-target effects Developing simulation-informed architectures that integrate molecular dynamics and physical priors Engineering scalable training pipelines across multi-cloud and distributed infra Collaborating with domain experts in chemistry, biology, and pharmacology to ground models in real-world constraints Prototyping and evaluating novel architectures for reasoning over multimodal biological data
Ideal Background: Experience building or publishing on generative models for molecules, proteins, or materials Familiarity with molecular dynamics simulations or physics-informed ML Expertise in graph neural networks, transformers, or diffusion models Track record of shipping ML systems in production or research environments Interest in model interpretability, uncertainty, and scientific grounding Comfort working across disciplines and translating abstract scientific goals into concrete ML systems
Bonus: Experience with ADMET modeling, QSAR, or cheminformatics Familiarity with simulation engines (e.g. OpenMM, GROMACS) or quantum chemistry tools Prior startup or zero-to-one experience Contributions to open-source ML or scientific computing projects
Candidates from any biological or scientific background are encouraged to apply. No specific domain (e.g., ADMET, chemistry, etc.) is required. Intellectual honesty, engineering excellence, and scientific curiosity are valued above domain specialization.
Apply now, read the job details by scrolling down Double check you have the necessary skills before sending an application.
Focus Areas: Designing generative models (e.g. diffusion, flow matching, graph-based) for molecular structure and dynamics Building predictive systems for ADMET properties, drug-likeness, and off-target effects Developing simulation-informed architectures that integrate molecular dynamics and physical priors Engineering scalable training pipelines across multi-cloud and distributed infra Collaborating with domain experts in chemistry, biology, and pharmacology to ground models in real-world constraints Prototyping and evaluating novel architectures for reasoning over multimodal biological data
Ideal Background: Experience building or publishing on generative models for molecules, proteins, or materials Familiarity with molecular dynamics simulations or physics-informed ML Expertise in graph neural networks, transformers, or diffusion models Track record of shipping ML systems in production or research environments Interest in model interpretability, uncertainty, and scientific grounding Comfort working across disciplines and translating abstract scientific goals into concrete ML systems
Bonus: Experience with ADMET modeling, QSAR, or cheminformatics Familiarity with simulation engines (e.g. OpenMM, GROMACS) or quantum chemistry tools Prior startup or zero-to-one experience Contributions to open-source ML or scientific computing projects
Candidates from any biological or scientific background are encouraged to apply. No specific domain (e.g., ADMET, chemistry, etc.) is required. Intellectual honesty, engineering excellence, and scientific curiosity are valued above domain specialization.