G
Research Engineer, AGI Safety and Alignment, DeepMind
Google
San Francisco, CA, USA; Mountain View, CA, USA
🏢 Офис
Middle
Полная занятость
Продуктовая
Tech
Описание вакансии
Обязанности
•Apply alignment techniques to frontier models
•Collaborate with product teams to adopt research
•Develop AGI control systems
•Implement AGI control systems in production
•Research alignment methods
•Research model interpretability techniques
•Study alignment failures
Технологии: Adversarial Robustness, Fine Tuning, Human Feedback, Language Models, Language Processing, Large Language Models, Learning from Human Feedback, Machine Learning, Model Interpretability, Natural Language, Natural Language Processing, Reinforcement Learning, Reinforcement Learning from Human Feedback, Supervised Fine Tuning