Research Scientist, Memory, Reasoning and Continual Learning, DeepMind
Core
Conducting fundamental research in AI to advance long-context attention mechanisms, retrieval-augmented generation, continual learning architectures, and multi-step reasoning frameworks.
Role type
Senior IC research scientist (memory, reasoning, continual learning)
Builds
Novel AI architectures, evaluation benchmarks, and high-capacity context window infrastructure
Domain
Artificial Intelligence, Machine Learning, Cognitive Science
Deliverable
production ML models
Required skills
Deep Learning, Reinforcement Learning, Natural Language Processing, Continual/Lifelong Learning, Algorithms Design, Experimentation, Python, JAX, TensorFlow, PyTorch
Preferred skills
Large Language Models, Autonomous Agents, Long-context Memory Systems, Benchmark Design, Computational Neuroscience, Distributed Training, Infrastructure for High-Capacity Context Windows
Technologies
JAX, TensorFlow, PyTorch
Responsibilities
Initiate and lead novel research directions in long-context attention and reasoning; Design and execute end-to-end experiments to align model-based agents and mitigate catastrophic forgetting; Develop evaluations and benchmarks for long-horizon memory and complex planning; Build and improve infrastructure for high-capacity context windows and continuous training pipelines; Communicate research findings through plots, writeups, and paper-ready narratives
Seniority
Senior, hands-on IC
