Member Of Technical Staff - Pre-training Infra

Reflection AI

San Francisco, United States
On-site
Distributed training systems
Large machine learning models
Megatron deepspeed
Reflection’s mission is to build open superintelligence and make it accessible to all

Job Summary

  • Reflection’s mission is to build open superintelligence and make it accessible to all.
  • Build and scale distributed training systems that power frontier model pre-training.
  • Top-tier compensation: Salary and equity structured to recognize and retain the best talent globally.

Matching Summary

Reflection’s mission is to build open superintelligence and make it accessible to all.

Skills & Requirements

Must-have

  • distributed training systems
  • large machine learning models
  • Megatron DeepSpeed
  • model parallelism strategies
  • training throughput optimization
  • GPU utilization optimization
  • NCCL performance tuning
  • large-scale datasets

Nice-to-have

  • open superintelligence mission
  • talent-dense team
  • rapid experimentation systems
  • production-ready training systems

Key Requirements

  • Experience building or operating distributed training systems
  • Experience with modern distributed training frameworks
  • Familiarity with large-scale model parallelism strategies
  • Experience optimizing training throughput and GPU utilization
  • Familiarity with GPU communication libraries
  • Experience working closely with ML researchers
  • Strong debugging skills
  • Experience working with large datasets

Work Rights

Not specified

Tailored Resume

Cover Letter