
Research Engineer, Pretraining Scaling - London
AI
Anthropic is hiring a Research Engineer for its ML Performance and Scaling team in London to train production pretrained models reliably, efficiently, and at scale. The role spans performance optimization, hardware debugging, experimental design, and launch coordination, requiring hands-on LLM training experience or deep expertise with JAX, TPU, PyTorch, or large-scale distributed systems.
What you'll do
- Own critical aspects of the production pretraining pipeline, including model operations, performance optimization, observability, and reliability
- Debug and resolve complex issues across the full stack—from hardware errors and networking to training dynamics and evaluation infrastructure
- Design and run experiments to improve training efficiency, reduce step time, increase uptime, and enhance model performance
What you need
Must-have
JAXTPUPyTorchLarge-scale distributed systemsLarge language model trainingPerformance optimizationObservabilityEvaluation infrastructureDebugging
Nice-to-have
open_lmllm-foundrymesh-transformer-jaxScaling lawsML systems researchProduction ML systemsObservability tools
Soft skills
CommunicationCollaborationProblem solvingComfort with ambiguityAdaptabilityPassion for the craft
Research EngineerAI ResearchPretrainingMachine LearningLLM TrainingJAXTPUPyTorchLondonOn-site