> Markdown version of [/jobs/ext/4649101-rl-infrastructure-engineer](https://www.wearedevelopers.com/jobs/ext/4649101-rl-infrastructure-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# RL Infrastructure Engineer

- **Company:** Collision Partners, LLC
- **Location:** San Francisco, CA, United States
- **Contract:** Permanent contract
- **Skills:** Systems Engineering, Computer Clusters, Nvidia CUDA, Software Debugging, Distributed Computing Environment, Python (Programming Language), Reinforcement Learning, Pytorch, Large Language Models, Parallel Computation, SGLang, Machine Learning Operations
- **Published:** November 9, 2026
- **Apply:** https://www.juju.com/job/28_25197_848bd03cc9327663f80d40e6b136dfed

## About the Role

* Strong Python and PyTorch skills, with experience building distributed training or inference infrastructure.
 * Practical understanding of LLM reinforcement learning, including rollout policies, reward computation, and policy updates.
 * Experience profiling GPU workloads and resolving performance, correctness, or reliability issues in complex ML systems.
 * Ability to investigate problems across research and infrastructure code and validate changes on real training runs.
 * Experience developing RL training pipelines, particularly asynchronous or off-policy systems.
 * Work on agent environments, sandboxed execution, reward models, or verification pipelines.
 * Experience extending SGLang, vLLM, or comparable inference runtimes.
 * Experience with FSDP, tensor or pipeline parallelism, NCCL, CUDA, or Triton.

## Description

The role combines systems engineering with a practical understanding of RL. Infrastructure problems can change learning behavior: delayed weight updates, differences between trainer and inference log probabilities, or off-policy drift can resemble problems with the training method itself. You'll work closely with researchers to understand these interactions and improve performance without compromising training correctness., * Training and rollout systems. Coordinate trainers, rollout workers, environment execution, and reward computation across GPU clusters. Integrate SGLang for high-throughput sampling, improve trainer-to-inference weight synchronization, and support asynchronous rollouts and off-policy training.

 * Performance at scale. Improve distributed training through FSDP, tensor and pipeline parallelism, and efficient NCCL communication. Profile bottlenecks across compute, memory, and networking, and make targeted runtime or CUDA/Triton kernel improvements where they meaningfully improve training efficiency.
 * Training correctness and reliability. Develop monitoring and diagnostics that catch stale weights, trainer-inference log-probability mismatches, and off-policy drift during long runs. Build reproducible debugging workflows and recovery mechanisms, and use controlled comparisons to understand how infrastructure changes affect learning.