← All roles

Member of Technical Staff, Pretraining Systems

Magic· Pretraining Systems
Apply on company site
Hybrid (San Francisco)$275k - $550kfull-timePosted 7mo ago

About the role

Research Engineer to design and operate distributed infrastructure for large-scale pretraining of long-context models, focusing on scalability, reliability, and reproducibility across massive GPU clusters.

Responsibilities

  • Scale distributed training across large GPU clusters (data, tensor, pipeline parallelism)
  • Optimize communication patterns and gradient synchronization
  • Improve checkpointing, fault tolerance, and job recovery systems
  • Profile and eliminate performance bottlenecks across compute, networking, and storage
  • Improve experiment reproducibility and orchestration workflows
  • Increase hardware utilization and training throughput
  • Collaborate with Kernels and Research to align model architecture with systems realities

Qualifications

  • Strong software engineering and distributed systems fundamentals
  • Experience training large models in multi-node GPU environments
  • Deep understanding of parallelism strategies and performance trade-offs
  • Experience debugging cross-layer issues in production ML systems
  • Strong ownership mindset and ability to operate critical infrastructure
  • Track record of improving performance or reliability of large-scale systems

Skills

Required:Distributed systemsLarge-scale model trainingParallelism strategies
Nice to have:Production ML systems debugging

Benefits

  • Annual salary range: $275K - $550K
  • Equity is a significant part of total compensation, in addition to salary
  • 401(k) plan with 6% salary matching
  • Generous health, dental and vision insurance for you and your dependents
  • Unlimited paid time off
  • Visa sponsorship and relocation stipend to bring you to SF, if possible
  • A small, fast-paced, highly focused team
Last updated Oct 21, 2026. Interested? Apply on the company site.
Magic logo

About Magic

building safe AGI to accelerate humanity’s progress on the world’s most important problems.

Magic is an AI company that is working toward building safe AGI to accelerate humanity’s progress on the world’s most important problems.

AGIAI researchArtificial IntelligenceModelsResearchliquid-ai-recruiting-target
Company profileWebsite

More open roles at Magic