Apply machine learning evaluation and production coding to chemical, biological, radiological, nuclear, and explosives AI safety at Google.

Google’s CBRNE team needs a research engineer for applied research, safeguards, and frontier model evaluations. You will work with biologists, chemists, and physicists while developing systems that address catastrophic harm risks. The work includes sensitive data management, strict release timelines, and on-call responsibilities.

Key Responsibilities:

- Build safeguards and test them through automated and manual red teaming
- Measure safeguards’ costs to legitimate scientific research
- Design evaluations for CBRNE and agentic systems risks
- Architect reliability for safeguards and incident response




- Communicate uncertainty and negative findings to technical leadership

Requirements:

- Bachelor degree or equivalent practical experience in a relevant field
- Eight years writing code in shared version-controlled codebases
- Three years training, fine-tuning, or evaluating machine learning models
- Experience building classifiers or evaluation infrastructure for language models
- Experience with CBRNE security, export control, or adversarial work

Use machine learning, software engineering, and scientific risk evaluation to strengthen Google’s frontier AI safeguards.

📌 Chemical Biological Nuclear AI Engineer (Toronto)
🏢 Google
📍 Toronto