Senior GPU ML Infra Engineer — Mid-Training & Inference

Reflection AI; San Francisco (CA); On-site; USD 120,000 - 160,000; Full time

Benefits offered by this job

Top-tier compensation

Comprehensive medical, dental, vision, life, and disability insurance

Fully paid parental leave for all new parents

Paid time off and relocation support

Daily meals and team celebrations

Job summary

A cutting-edge AI technology company based in San Francisco is seeking a specialist to design and operate large-scale GPU infrastructure. This role requires expertise in deploying GPU systems for high-throughput inference and model performance optimization. The ideal candidate will have hands-on experience with modern inference frameworks and a solid understanding of reinforcement learning technologies. Comprehensive healthcare benefits, parental leave, and daily meals are provided, along with competitive salary and equity packages.

Qualifications

• Experience with high-throughput model inference and mid-training workloads.
• Strong technical skills in optimizing GPU utilization for inference.
• Familiarity with large-scale distributed systems and RL workflows.

Responsibilities

• Design and build GPU infrastructure for model inference.
• Develop synthetic data generation systems and RL pipelines.
• Optimize latency and throughput for large language models.
• Support distributed RL workloads and model evaluation.

Skills

Experience deploying and operating large-scale GPU systems

Hands-on experience building and running production infrastructure

Understanding GPU performance characteristics

Experience with modern inference frameworks

Familiarity with distributed reinforcement learning infrastructure

Experience optimizing throughput for model execution workloads

Experience with GPU kernels and performance optimization

Familiarity with infrastructure for synthetic data pipelines

Debugging performance issues across GPU and distributed layers

Job description

A cutting-edge AI technology company based in San Francisco is seeking a specialist to design and operate large-scale GPU infrastructure. This role requires expertise in deploying GPU systems for high-throughput inference and model performance optimization. The ideal candidate will have hands-on experience with modern inference frameworks and a solid understanding of reinforcement learning technologies. Comprehensive healthcare benefits, parental leave, and daily meals are provided, along with competitive salary and equity packages.