Trust and Safety Intel Analyst, Model Behavior and Alignment
Washington D.C., DC, USA · Full-time · Posted Oct 8, 2026
- Experience
- 7 years in one or more of the listed fields.
- Education
- Bachelor's degree or equivalent practical experience.
About the job
Owns intelligence analysis of model behavior risks across Google products, turning incidents and other signals into risk assessments, mitigations, and launch and safety decisions.
What you'll do
- Maintain a company-wide view of material risks and coordinate cross-functional reviews of signals.
- Produce quick-turn and strategic assessments, distinguishing genuine misalignment from misattributed failures or hype.
- Develop threat models and mitigations, track residual risks, and build AI-assisted workflows for research monitoring and signal triage.
What they ask for
- Bachelor's degree or equivalent practical experience.
- 7 years of experience in one or more of: AI safety or alignment research, AI policy or governance, model evaluations or red teaming, trust and safety, intelligence or threat analysis, or risk analysis.
- Experience critically assessing technical AI research or evaluation results and communicating findings to non-specialists.
Nice to have
- Experience in AI safety, alignment, model evaluations, and AI governance.
- AI evaluation across modalities, Python/SQL scripting, or familiarity with misalignment research, safety policy, or threat modeling.
Pay
$141000 - $205000 USD base salary
Tools
Python, SQL
AI focus
Assess AI model behavior risks, including misalignment across chatbots, multimodal assistants, agents, and robotics; use AI-assisted workflows for research monitoring and signal triage.
Model-written summary · the original posting is authoritative