Machine Learning Engineer

EPM Scientific are supporting a growing biotech focused on advancing computational protein engineering. The team combines large‑scale experimental assays with modern machine learning to design proteins with specific structural and functional properties. Their work centers on creating models that can propose high‑value variants and rapidly validate them through a tight lab‑in‑the‑loop cycle.

This role is ideal for someone who wants to work at the intersection of sequence modeling, structural biology, and experimental feedback, and who wants their modeling work to directly influence real design decisions.


Role Overview

You will develop and refine machine learning models that learn sequence-structure-function relationships and guide the design of optimized proteins. You will work closely with experimental scientists to interpret assay results and incorporate them into the next round of modeling.



Your work will include:

• Building generative and predictive models for protein sequences and structures using approaches such as transformers, diffusion models, graph networks, and continuous‑time architectures.
• Integrating sequence, structural, evolutionary, and assay‑derived data into unified modeling frameworks.
• Modeling mutational effects and fitness landscapes to identify promising variants for testing.
• Running iterative design cycles that incorporate experimental measurements and use them to update or guide the model.
• Collaborating with protein engineers, structural biologists, and assay teams to translate model outputs into practical designs.

What You Bring:

• PhD or equivalent experience in Machine Learning, Computational Biology, Bioinformatics, Structural Biology, or a related discipline.
• Hands‑on experience developing generative or sequence‑modeling architectures.
• Familiarity with protein representation learning, structure prediction, or modeling mutational effects.
• Proficiency in Python and experience with PyTorch or JAX.
• Ability to work closely with experimental collaborators and adapt models based on real‑world constraints.
• Interest in building systems that produce practical design candidates, not just conceptual models.