jobs

Research Engineer, AGI Safety and Alignment, DeepMind

The Artificial General Intelligence (AGI) Safety and Alignment Team (ASAT) aims to reduce existential and catastrophic risk from AGI and eventually Artificial Superintelligence (ASI). We research novel techniques and work with the rest of G

In inglese

Company

Google

Location

London, UK

Status

Open

Posted

2026-08-25T17:40:05+00:00

Google DeepMind's AGI Safety and Alignment Team in London hires research engineers and scientists on-site, as Research Scientists or Software Engineers. Work covers alignment methods, alignment failures, AGI control systems and interpretability research. Priority areas include deep alignment, alignment stress testing, language model interpretability, agent control and amplified oversight. Candidates need a bachelor's degree in computer science or a related software engineering field, three years of software development, ML engineering or ML research experience, and work with research teams; training large models with supervised finetuning or RLHF is preferred.