Cyber AI Safety Engineer - Post-Training & RL

Google LLC; Mountain View, New York (CA, NY); On-site; USD 147,000 - 210,000; Full time

Benefits offered by this job

Equity

Bonus target

Benefits

Job summary

Google is seeking a Research Engineer for the Gemini Cyber Post-Training team. The role focuses on advancing vulnerability discovery through post-training recipes, RL datasets, and scalable evaluation pipelines.

You will collaborate with DeepMind and internal research programs to push AI safety and security forward. Ideal candidates have a PhD or equivalent experience, at least two years in RL, and publications in vulnerability research.

Qualifications

• PhD or equivalent practical experience in CS or related field.
• 2 years of RL experience in industry or academia.
• Publications in vulnerability research (auditing/reverse engineering/exploitation).

Responsibilities

• Build datasets (RL environments, SFT data) and iterate on post-training recipes.
• Debug and analyze RL training runs to identify bottlenecks and optimize training.
• Build evaluation pipelines, benchmarks, and leaderboards to measure agent performance.

Skills

Reinforcement Learning

Machine Learning

Publications in vulnerability research

Education

PhD in Computer Science or related field, or equivalent experience

Job description

Google is seeking a Research Engineer for the Gemini Cyber Post-Training team. The role focuses on advancing vulnerability discovery through post-training recipes, RL datasets, and scalable evaluation pipelines.

You will collaborate with DeepMind and internal research programs to push AI safety and security forward. Ideal candidates have a PhD or equivalent experience, at least two years in RL, and publications in vulnerability research.