Frontier RL Systems Engineer

Prime Intellect; San Francisco, Northern (CA, KY); Hybrid; USD 150,000 - 350,000; Full time

Benefits offered by this job

Visa sponsorship

Relocation assistance

Remote work option

Job summary

Prime Intellect is building the open frontier AI stack, delivering a full-stack system for frontier-scale model training and evaluation. You will contribute to scalable RL and distributed training infrastructure, focusing on performance, memory, and communication optimizations, and work closely with researchers and engineers.

The role offers flexible work arrangements with remote or SF office options, visa sponsorship, relocation support, and a highly technical, ownership-driven team environment.

Qualifications

• Strong systems engineering experience in AI/ML infrastructure for large-scale training or inference.
• Experience with PyTorch Distributed, DeepSpeed, FSDP, Megatron, vLLM, Ray or related tooling.
• Ability to optimize training performance across kernels, memory movement, and communication overhead.
• Hands-on experience with data, tensor, and pipeline parallelism at scale.
• Deep understanding of GPU architectures and performance debugging.
• Ability to diagnose bottlenecks and drive principled improvements across the stack.

Responsibilities

• Build and optimize the systems infrastructure behind large-scale RL and distributed training workloads by contributing to our prime-rl framework.
• Improve end-to-end training efficiency across compute, memory, networking, and scheduling layers.
• Design and implement low-level performance optimizations, including kernels, communication paths, and runtime improvements.
• Work on distributed training systems spanning data, tensor, and pipeline parallel workloads.
• Help shape the architecture of our RL training stack, including async rollout and post-training systems.
• Contribute to open-source libraries and internal infrastructure used for frontier-scale model training.
• Collaborate with researchers and infra engineers to translate bottlenecks into concrete improvements.
• Stay at the frontier of training systems, inference systems, compiler/runtime tooling, and hardware-aware optimization techniques.

Skills

Systems engineering

PyTorch & distributed training

Performance optimization

Large-scale training

GPU architecture

Bottleneck identification

Adaptability

Tools

CUDA kernels

Triton kernels

Open-source ML infra tools

Job description

Prime Intellect is building the open frontier AI stack, delivering a full-stack system for frontier-scale model training and evaluation. You will contribute to scalable RL and distributed training infrastructure, focusing on performance, memory, and communication optimizations, and work closely with researchers and engineers.

The role offers flexible work arrangements with remote or SF office options, visa sponsorship, relocation support, and a highly technical, ownership-driven team environment.