EmergingGenomics, multi-omics and biological data

Biological Foundation Model Researcher

Also called Biology Foundation Model Scientist

A Biological Foundation Model Researcher builds broad models over biological sequences or molecular measurements, then shows they work on real biology. It matters because a model that only scores well on benchmarks will not help anyone make a discovery.

What they do

The work starts with the data. They choose which sequences or measurements to model, clean and organise them, and decide how to split training from evaluation so that related biology does not leak across the line. They design and train sequence or multi-omics models, then test what those models have learned. Does the representation transfer to tasks it was never trained on? Does it reflect real biology, or just dataset quirks? They work closely with wet-lab scientists, read papers critically, and write down what failed as well as what worked.

What a strong hire delivers

In the first months, a strong hire delivers a clear evaluation set built on independent biological tasks, so the company can tell progress from noise. They produce a working baseline model with documented transfer results. They flag where the data is weak, biased or sensitive. They also give the team an honest view of which biological questions the model can help with now and which it cannot.

How you can tell they are good

Ask them to walk through a model they built and how they knew it captured biology rather than artefacts. Strong people talk about leakage, held-out families, and experiments that could have shown them wrong. Ask what they would test first with a new dataset. Be wary of anyone who cites only benchmark scores or cannot explain a failed result.

Why it is hard to fill

This is an emerging role, often hidden inside titles like computational biologist, ML scientist or research scientist. Few people are strong in both modern deep learning and real biological judgement. Most lean one way. Candidates who have trained large models and also understand assay noise, population structure and experimental design are scarce, and they are usually well placed where they are.

Where to find them

Look at computational biology and bioinformatics groups, genomics and multi-omics teams in biotech, and ML researchers who have moved into biological problems. Academic labs working on sequence models are a source, as are techbio companies and the biology teams inside larger AI organisations. Titles vary widely: research scientist, applied scientist, computational geneticist, or machine learning engineer in biology.

Indicative pay

Not enough published evidence yet to give a fair range for this role, which is common for work this new. We benchmark it for you as part of a search.

Interview questions for a Biological Foundation Model Researcher

  1. Describe a model you trained on biological sequences or omics data. How did you decide it had learned something real?

    Strong answers describe independent tests, held-out biology, and ideally follow-up experiments. They name what would have disproved the claim. Worry if the evidence is only a leaderboard score or if they cannot say what the model actually represents.

  2. How would you split data for a sequence model so that evaluation is not inflated by related sequences?

    Good answers cover homology, species or family level holdouts, batch and cohort effects, and checking for leakage. They explain the trade-offs. Worry if they default to random splits or have never considered redundancy in biological data.

4 more questions, with what a strong answer covers, are in the Role Kit.

Job description you can post

We are building broad models over biological sequences and molecular measurements, and we need a researcher who can show they work on real biology. You will train models, design the tests that judge them, and work with scientists who run the experiments.

What you will do

  • Develop and train models over biological sequences or multi-omics measurements
  • Design transfer tests on independent biological tasks
  • Build evaluation sets that avoid leakage between related data
  • Work with experimental scientists to validate model predictions
  • Build and maintain data pipelines over omics databases
  • Document results, including failures, for the wider team

What you will bring

  • Shown work training models on biological sequences or molecular data
  • Clear grasp of probability and statistics, applied to noisy biological data
  • Evidence of testing biological validity, not only benchmark performance
  • Experience building reproducible bioinformatics or data pipelines
  • Ability to explain results to both ML and biology colleagues
  • Care with sensitive data and privacy constraints

Nice to have

  • Experience with biomolecular modelling
  • Published or open work on biological representation learning
  • Experience collaborating on wet-lab validation

Free to use and adapt. If it saves you time, a link back to this page is appreciated.

Skills that matter

  • Biomolecular modellingAdvancedRepresent and predict molecular structure and interactions
  • Data pipelinesAdvancedBuild reliable ingestion and transformation systems
  • Probability and statisticsWorkingQuantify variability and statistical evidence
  • Privacy engineeringWorkingLimit disclosure and unauthorised personal-data use

Other roles in genomics, multi-omics and biological data

Hiring nowGenomics Machine Learning ScientistA Genomics Machine Learning Scientist builds models of genomic variation and sequence function, and shows they hold up across populations and…Hiring nowAI Bioinformatics EngineerAn AI Bioinformatics Engineer builds the pipelines that turn raw biological data into clean, trustworthy inputs for models. Without that work, a…EmergingMulti-Omics AI ScientistA Multi-Omics AI Scientist builds models that combine several types of biological measurement, such as genomic, transcriptomic and proteomic data…

Related roles in other families

Hiring nowCredit Risk AI ScientistFinancial services risk and compliance AIA Credit Risk AI Scientist builds credit models and proves they hold up. They show the models predict well, stay calibrated and treat groups fairly…Hiring nowFinancial Crime AI SpecialistFinancial services risk and compliance AIA Financial Crime AI Specialist builds and evaluates models that surface suspicious activity inside governed investigation workflows. They help a…Hiring nowAI Financial Model ValidatorFinancial services risk and compliance AIAn AI Financial Model Validator independently challenges the financial AI models a bank or payments company relies on. They find weaknesses before…EmergingBanking AI Governance SpecialistFinancial services risk and compliance AIA Banking AI Governance Specialist sets up and runs the controls around AI in banking and payment workflows. They help a company show regulators…

How we wrote this

Role definitions are our own editorial work, checked against occupational classifications, standards, research and employer pages. Titles vary by company, and this is a description of the work, not a job advert.

← Genomics, multi-omics and biological dataAll new job families

Start a search

Hiring a Biological Foundation Model Researcher?

Tell us the work. We find the people already doing it, whatever their job title says, and you meet them already interviewed for your role.