ML Research Scientist - De Novo Design

Merge Labs is a frontier research lab with the mission of bridging biological and artificial intelligence to maximize human ability, agency and experience. We’re pursuing this goal by developing fundamentally new approaches to brain-computer interfaces that interact with the brain at high bandwidth, integrate with advanced AI, and are ultimately safe and accessible for anyone to use.

About the team:

Merge is building the next generation of brain-computer interfaces by combining recent advances in synthetic biology, neuroscience, AI, and non-invasive imaging. To support this mission, we are building a cross-functional data-science group which sits at the intersection of computational modeling, neuroscience, and biomolecular engineering. This group collaborates extensively with wet-lab scientists, automation engineers, and data engineers to create ML frameworks that accelerate molecule discovery and device optimization.

About the role:

We’re hiring a Senior / Principal ML Scientist to design and scale de novo design frameworks that guide molecular engineering campaigns through iterative design–build–test–learn (DBTL) cycles. Starting from a blank slate, you’ll first architect the company’s closed-loop optimization backbone—building the data and modeling foundations that connect experiments to these ML frameworks. Over time, you’ll help translate these prototypes into production pipelines that measurably improve experimental throughput and discovery success across multiple biomolecular and neuroengineering verticals.

In this role, you will:

• Build the scientific and engineering scaffolding in collaboration with data-engineering & MLOps for de novo design and closed-loop optimization, including data ingestion, ML modeling, and library design.

• Collaborate with wet-lab scientists to define tractable optimization objectives and encode domain specific priors and constraints.

• Prototype de novo design frameworks using internal and public datasets; benchmark and validate model performance.

• Integrate ML models with experimental data streams and serve to non-domain experts for model democratization.

• Extend ML frameworks to handle multi-objective or constrained optimization problems.

• Stay up-to-date with the latest research in de novo design, and prototype novel algorithms that can be deployed to improve the company’s discovery or development workflows.

• Contribute to the long-term research roadmap and serve as a thought-leader for scientists.

You might thrive in this role if you have:

• Strong grounding in SSMs, LLMs, SE(3)-equivariance, Flow-matching.

• Working knowledge of transfer-learning strategies.

• Proficiency in Python / PyTorch / Jax and comfort writing clean, reproducible production grade code.

• Experience bridging machine learning and experimental science – working with sparse, noisy, and or high-cost data.

• A collaborative, systems-level mindset.

• Nice to have: familiarity with neuroscience.

If you're excited about this role but don't meet every qualification, please apply. As we build, we're hiring for complementary strengths to form a high-impact team.

For more information about hiring at Merge, please visit our Hiring FAQ

Merge Labs does not discriminate on the basis of race, color, religion, national origin, age, sex, sexual orientation, gender, gender identity, gender expression, marital status, physical or mental disability, medical condition, genetic information, family status, ancestry, citizenship, U.S. military (state and federal) and veteran status, or any other legally protected status. It is our intention that all applicants be given equal opportunity and that selection decisions are based on job related factors. We are an equal opportunity employer.

Pursuant to the San Francisco Fair Chance Ordinance, we will consider for employment qualified applicants with arrest and conviction records.

We are committed to providing reasonable accommodations to applicants with disabilities, and requests can be made by emailing [email protected].