Chemist, DeepMind
New York, NY, USA +1 · Full-time · Posted Oct 7, 2026
- Location
- New York, NY, USA
Mountain View, CA, USA - Experience
- 2 years of post-doctoral, industry, or defense experience.
- Education
- PhD in synthetic or computational chemistry, or equivalent practical experience.
About the job
Evaluates and mitigates chemistry-related safety risks in frontier AI models while minimizing harm to legitimate scientific work.
What you'll do
- Design, run, and review computational chemical evaluations, including synthesis planning, predictive toxicology, and molecular docking.
- Assess synthetic chemistry model outputs and their potential to increase a threat actor’s capabilities.
- Recommend safety mitigations, update harm and threat frameworks, and communicate findings to technical and executive audiences.
What they ask for
- PhD in synthetic or computational chemistry, or equivalent practical experience.
- Wet-lab bench experience designing, executing, and troubleshooting multi-step syntheses.
- Experience in cheminformatics and digital drug discovery, including Python, dataset analysis, and binding-affinity prediction.
- Experience with computer-aided synthesis planning, retrosynthesis, 3D structure-based modeling, and Scientific AIs or applying LLMs to scientific problems.
Nice to have
- Experience with dual-use chemistry, the Chemical Weapons Convention (CWC), and chemical control treaties.
- Understanding of LLM principles, LLM mitigation strategies, and frontier AI model evaluation methodologies.
Pay
$174000 - $252000 USD base salary
Tools
Python, RDKit, QSAR, protein-ligand co-folding, FEP
AI focus
Evaluates frontier AI models’ chemistry capabilities and risks, including agentic capability testing, and recommends mitigations to reduce dual-use chemical harm.
Model-written summary · the original posting is authoritative