AI Safety & Alignment Research Scientist (GenAI)

DeepMind Technologies Limited; Mountain View (CA); Ibrido; USD 207.000 - 300.000; Tempo pieno

Descrizione del lavoro

DeepMind Technologies Limited’s Alignment and Reliability division seeks a Research Scientist/Engineer to advance novel heuristic and statistical mechanisms for safe, reliable AI systems. You will own end-to-end experiments from conception to live release within a fast, collaborative team.

We pursue ethical integrity across flagship interfaces and APIs, with a strong emphasis on safety, user-benefit, and scientific discovery, including published work and open collaboration.

Competenze

• PhD or equivalent practical experience in CS or related field.
• 2 years of experience in Large Language Model safety or security.

Mansioni

• Drive safety and security mitigations for frontier GenAI models at scale in partnership with product areas.
• Develop red and blue teaming methods for GenAI models including text-to-text, multimodal, and agentic capabilities.
• Explore data, reasoning and algorithmic solutions to ensure Gemini Models are safe and helpful for everyone.
• Improve Gemini's adversarial robustness focusing on high-stakes abuse risks in agentic settings.
• Develop and execute experimental plans to address known gaps or create new capabilities.

Conoscenze

LLM safety

Agentic workflows

Synthetic data pipelines

Research to product

Applied research leadership

Publications at NeurIPS/ICML/ICLR

Formazione

PhD in Computer Science or related field

Descrizione del lavoro

DeepMind Technologies Limited’s Alignment and Reliability division seeks a Research Scientist/Engineer to advance novel heuristic and statistical mechanisms for safe, reliable AI systems. You will own end-to-end experiments from conception to live release within a fast, collaborative team.

We pursue ethical integrity across flagship interfaces and APIs, with a strong emphasis on safety, user-benefit, and scientific discovery, including published work and open collaboration.