Aleph Alpha GmbH 5 дней назад

AI Researcher - Pre-training

  • Heidelberg
  • Гибрид
LLMMachine LearningOptimizationTransformerExperiment Designscaling lawKernel OptimizationattentionMegatron-LMPipeline Parallelismtensor parallelismTorchTitan

Описание

Responsibilities

  • Design and run foundation-model experiments
  • Improve training recipes
  • Support large-scale training runs

Requirements

  • machine learning
  • TorchTitan
  • attention
  • LLM
  • Megatron-LM
  • experiment design
  • tensor parallelism
  • optimization
  • kernel optimization
  • Pipeline Parallelism
  • transformer
  • scaling law
  • DeepSpeed
  • PyTorch
  • reproducibility
  • probability
  • Python
  • Distributed training
  • expert parallelism
  • data parallelism

We offer

  • hybrid work mode

Hiring process

  • Reach out to ••• for Product & Engineering Leadership and ••• for Backend Engineering and Archictecture.