Senior Research Scientist AI Safety and Biosecurity
Basecamp Research is an AI-biology company building the BaseData genomic dataset and EDEN biological foundation models for therapeutic design.
About Basecamp Research
The company uses proprietary evolutionary sequence data to train EDEN foundation models, including models that can be accessed through Claude for antibiotic-peptide design and immunogenicity prediction. Its therapeutic platform also includes AI-Programmable Gene Insertion (aiPGI).
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will advance AI safety and biosecurity research for biological AI systems. You will design evaluations, red-team models, develop biologically grounded threat models, investigate model behaviour and generalisation, and test safeguards across training, screening, behaviour, and access. You will lead projects from research questions through experiments and reusable methods, communicate findings and limitations, and help determine research priorities. You may publish research and collaborate with the wider AI safety and biosecurity community.
Requirements
- Demonstrated original research in machine learning, AI safety, computational biology, biosecurity, or a closely related field
- Deep expertise in model evaluation, interpretability, robustness, safeguards, biological modelling, or quantitative biological-risk assessment
- Understanding of modern machine learning
- Ability to write code, run experiments, and analyse results
- PhD in a relevant field or equivalent research record
- Experience leading substantial research projects
- Ability to work closely with domain experts and assess biological evidence
- Ability to design controlled experiments, evaluate alternative explanations, and communicate findings and limitations clearly
Responsibilities
- Design evaluations and red-teaming studies for biological AI systems
- Develop biology-grounded threat models
- Work with experimental scientists to assess real-world capabilities and risks
- Investigate model behaviour, representations, and generalisation
- Develop and test safeguards across training, model behaviour, design screening, and access
- Measure safeguard effectiveness and effects on scientific work
- Lead research projects from initial questions through experiments and reusable methods
- Inform model development, publications, releases, and services
- Publish research where appropriate
- Collaborate with the AI safety and biosecurity community
Benefits
- Equity
- Private healthcare with no medical history exclusions
- Carrot Fertility with IVF stipend
- Salary sacrifice pension
- Bike to work scheme
- Life insurance
- Coaching
- Twice-yearly promotion opportunities