Research Engineer, Gemini Agent Safety Post Training
Google DeepMind · Mountain View, CA, USA
Frontier labonsite$174k–$252k
At a glance
Highlights
- Architect alignment mechanisms for Gemini agents operating at global scale.
- Pioneer safety alignment algorithms and dynamic evaluations for frontier models.
- Drive safe deployment of agents across high-impact production surfaces.
What they offer
- 15% bonus target
- equity
- benefits
Ideal candidate
ML researcher passionate about AI safety and alignment who thrives partnering across teams to ship reliable agents under tight timelines.
Requirements
- Bachelor's degree or equivalent practical experience
- 2 years of software development or ML research experience, or education equivalent credit
- Experience in machine learning algorithms, deep learning research, and large language modeling
- Experience in AI safety, alignment, and robust decision-making including SFT, RLHF, RLAIF, online distillation
- Experience training and fine-tuning large models at scale across distributed compute clusters
- Experience developing or red-teaming autonomous agents in interactive sandbox environments
- Expertise in designing and scaling automated evaluation frameworks, red-teaming pipelines, or dynamic benchmarks
Lab profile
About Google DeepMind
Google's AI lab building frontier models like Gemini to advance science and general intelligence.
What they build
- Gemini
- Gemma
- AlphaFold
Known for
- Frontier foundation models and GenAI
- Large-scale training with JAX, PyTorch, TensorFlow
- Model evaluation and deployment expertise
- Responsible frontier AI research
Key facts
- Founded in 2010
- Headquartered in London, UK
- Part of Google (Alphabet)
First seen · Last seen