Research Engineer, Agi Safety and Alignment, Deepmind

Google Google · Big Tech · London, United Kingdom

Research Engineer focused on AGI Safety and Alignment, aiming to reduce existential and catastrophic risk from AGI and ASI. The role involves researching novel alignment techniques, studying alignment failures, applying AGI-scalable alignment to frontier models, developing and implementing adversarially robust AGI control systems, researching interpretability techniques, and working with product teams for research adoption. The role also involves advising executive leadership on AI risks.

What you'd actually do

  1. Research new alignment methods, studying alignment failures, and applying AGI-scalable alignment techniques to frontier models.
  2. Develop adversarially robust AGI control systems and implement them in production.
  3. Research interpretability techniques to understand what AI systems are ‘thinking’.
  4. Work with product teams to ensure that our research is correctly adopted.

Skills

Required

  • software development
  • ML engineering
  • ML research
  • working with research teams

Nice to have

  • applied research to improve the safety and alignment of frontier AI systems
  • training large models (e.g., supervised finetuning, RLHF)

What the JD emphasized

  • existential and catastrophic risk
  • AGI-scalable alignment techniques
  • adversarially robust AGI control systems
  • interpretability techniques
  • frontier AI systems
  • training large models

Other signals

  • AGI Safety and Alignment
  • existential risk
  • frontier models
  • interpretability
  • control for agents
  • alignment techniques