Research Scientist, Efficient Long Context and Memory, Deepmind

Google Google · Big Tech · New York, NY +1

Research Scientist at Google DeepMind focused on advancing AI development, specifically in efficient long context and memory for AI agents. The role involves identifying weaknesses in current agent capabilities, establishing research roadmaps, and contributing to fundamental research through publications and collaborations. The position emphasizes scaling deep neural network architectures and potentially translating research into engineering systems or product applications.

What you'd actually do

  1. Identify fundamental weaknesses in current agent capabilities and establish research roadmaps and technical strategies to systematically solve them.
  2. Mentor, guide, and elevate junior scientists and engineers, fostering a culture of technical excellence and innovation.
  3. Guide ambitious team collaborations, aligning efforts across Google DeepMind to execute on shared, high-impact research goals.
  4. Report and present complex research breakthroughs clearly and efficiently. Act as a principal representative of our work internally and externally at global venues.
  5. Build, lead, and maintain collaborations with key external research labs and leading individuals in the field.

Skills

Required

  • PhD degree in Computer Science or related field
  • Post-PhD academic or industry research experience in AI
  • Scientific publication record

Nice to have

  • Mentoring researchers
  • Translating fundamental AI research into engineering systems or product applications
  • Scaling deep neural network architectures
  • Technical lead or principal investigator on large-scale ML projects
  • Long context and memory research
  • Agent capabilities research

What the JD emphasized

  • PhD degree in Computer Science, a related field, or equivalent practical experience.
  • Experience with post-PhD academic or industry research in the field of artificial intelligence (AI).
  • One or more scientific publication submission(s) for conferences, journals, or public repositories (NIPS, ICML, ACL, CVPR, etc.).
  • Demonstrated experience in mentoring researchers and successfully translating fundamental AI research into major engineering systems or product applications.
  • Deep, authoritative expertise in scaling deep neural network architectures.
  • A proven track record of acting as a technical lead or principal investigator on complex, large-scale machine learning projects.

Other signals

  • fundamental research
  • long context
  • memory
  • agent capabilities
  • scaling deep neural networks