Research Engineer, GenMedia Safety Evaluations and Threat Intelligence TLM, DeepMind

Google DeepMind’s GenMedia Safety team defines the safety and alignment behavior of our flagship image, video, audio, and real-time interactive generative models. Working across safety post-training, runtime guardrails, evaluation, and threat intelligence, we address multi-stage systems whose visual and audio outputs carry high real-world stakes. Our mission is to protect people and society from harm while preserving creative freedom, neutrality, and user delight. We are responsible for making sure that Google DeepMind’s generative media models are deployed with the best GenAI safety in the world.

As Technical Lead and Manager for GenMedia Safety Evaluations and Threat Intelligence, you will build and lead a team of researchers and engineers responsible for measuring, stress-testing, and anticipating risks across multimodal, multi-turn, and agentic generative media systems. You will own how we translate nuanced safety, alignment, and product goals into measurable quantities, and you will ultimately own the quality and trustworthiness of data across the entire GenMedia Safety team. Working collaboratively with peers across post-training, runtime safety, and production monitoring, you will set the strategy and architect the solution that lets us make data-driven decisions across the entire development and deployment life-cycle. Artificial intelligence will be one of humanity’s most transformative inventions. At Google DeepMind, we are a pioneering AI lab with exceptional interdisciplinary teams focused on advancing AI development to solve complex global challenges and accelerate high-quality product innovation for billions of users. We use our technologies for widespread public benefit and scientific discovery, ensuring safety and ethics are always our highest priority.

We are pushing the boundaries across multiple domains. Our global teams offer diverse learning opportunities and varied career pathways for those driven to achieve exceptional results through collective effort.

Individual pay is determined by factors including job-related skills, experience, and relevant education or training.

US: $262000 - $364000 (USD) + 25% bonus target + equity + benefits

Learn more about benefits at Google . Responsibilities

• Translate safety and alignment goals into operational metrics, calibrated multimodal automated judges, and robust evaluation benchmarks.

• Build production-log evaluations and creatively probe system blind spots to continuously update multimodal threat models.

• Evaluate the quality of steered responses across media types and influence system infrastructure to ensure comprehensive logging.

• Govern dataset quality across the team and verify post-launch metrics to fold real-world data back into regression suites.

• Manage and mentor a team of 5 to 10 researchers and engineers while fostering a culture of scientific excellence.

Minimum qualifications:

• PhD or equivalent practical research experience.

• 5 years of experience in a technical leadership role setting direction and guiding project workstreams.

• 5 years of experience in a people management, supervision, or team leadership role.

• 5 years of experience in machine learning, specifically working with LLMs, multimodal foundation models, or RL, along with safety and alignment specialization (e.g., model alignment, AI safety, or post-training techniques like supervised fine-tuning and reward modeling).

• One or more peer-reviewed scientific publications (e.g., NeurIPS, ICML, ICLR, CVPR, ACL, CoRL) or public preprints (arXiv).

Preferred qualifications:

• Master's degree or PhD in Computer Science, Artificial Intelligence, Robotics, or related technical fields.

• 5 years of experience driving cross-functional initiatives across research, engineering, infrastructure, and product stakeholders.

• 5 years of technical leadership experience leading engineering workstreams, setting multi-year technical goal, and mentoring staff engineers.

• Demonstrated research in LLM reasoning, multimodal systems, RL/agents, robotics, or scientific AI.