About the job
Join our dynamic Post-Training & Alignment Technical Program Management (TPgM) team at Google DeepMind, where we focus on driving operational design and execution within the post-training scope. As a Technical Program Manager specializing in AI Safety, you will play a pivotal role in ensuring the safety of our innovative Gemini models. This position is essential for coordinating model training, evaluation infrastructure, and managing an extensive portfolio of cutting-edge AGI Safety & Alignment research initiatives.
About Us
At Google DeepMind, we are dedicated to advancing the field of artificial intelligence for the benefit of humanity. Our diverse team of scientists, engineers, and machine learning experts collaborates to push the boundaries of AI technology, ensuring that safety and ethical considerations remain at the forefront of our work.
Your Role
- This is an impactful opportunity for a skilled Technical Program Manager focused on AI Safety. You will work closely with a globally distributed team based in Mountain View, New York, London, and Zurich, all united in the mission to develop, deploy, and research safe AGI. This role requires an individual comfortable with navigating the uncertainties of pioneering AI research while maintaining a fast-paced workflow. Your organizational skills will be critical in managing complex stakeholder relationships and fostering collaboration across Google DeepMind to achieve successful outcomes.
Key Responsibilities:
- Facilitate the end-to-end delivery of SFT and RL training data for AI Safety by establishing and maintaining essential structures (meetings, project plans) for timely delivery.
- Deeply analyze technical details, including loss functions and evaluations for safety, to ensure they accurately address the problems we aim to solve.
- Support a diverse array of work, from model development to long-term research in AGI Safety & Alignment.
- Enhance organizational efficiency by refining processes, optimizing execution, and implementing tool improvements as needed.
- Proactively identify and address blockers to quickly remove obstacles hindering research and engineering teams.
- Coordinate with various groups within Gemini, including Release teams and Model Behavior teams, to ensure alignment and efficiency.