Frontier AI Security Research Engineer

DeepMind Technologies Limited; Mountain View (CA); Hybrid; USD 207,000 - 300,000; Full time

Job summary

Google DeepMind in the United States (San Francisco, Mountain View, New York) seeks a research-focused Software Engineer to build defenses against misuse of frontier models and advance cyber security for Gemini. You will experiment, prototype, and deploy scalable solutions that protect our AI systems and users.

You will apply engineering, ML, RL, and cybersecurity knowledge to strengthen defenses and contribute to Gemini’s capabilities. Join a team advancing AI safely for billions of users.

Qualifications

• PhD in CS or related field or equivalent practical experience.
• At least 5 years in cybersecurity (red teaming, vulnerability discovery, etc.).
• At least 5 years developing agentic tools.
• At least 2 years of people management experience.

Responsibilities

• Improve adversarial evaluation techniques.
• Drive improvements for out-of-model defenses such as classifiers and introspection probes.
• Collaborate with safety engineering to develop advanced capabilities.
• Contribute to Gemini’s cyber defense capabilities.
• Understand path from research to production; impact Gemini releases and users.

Skills

Red Teaming

Vulnerability ID

Reverse Engineering

Agentic Tools Development

People Management

Education

PhD in Computer Science or related field

Job description

Google DeepMind in the United States (San Francisco, Mountain View, New York) seeks a research-focused Software Engineer to build defenses against misuse of frontier models and advance cyber security for Gemini. You will experiment, prototype, and deploy scalable solutions that protect our AI systems and users.

You will apply engineering, ML, RL, and cybersecurity knowledge to strengthen defenses and contribute to Gemini’s capabilities. Join a team advancing AI safely for billions of users.