Lead RL Infrastructure Engineer - Scale Training Systems

Speedrun Talent Network; Greater London; Hybrid; GBP 120,000 - 190,000; Full time

Job summary

Speedrun Talent Network seeks an expert to design, build, and scale distributed RL post-training systems for large-scale models. You will orchestrate trainer, rollout, environment, and reward workloads across thousands of GPUs and integrate inference engines such as vLLM and SGLang.

You will develop RL environments and reward infrastructures, verifiers, and evaluation harnesses, ensuring stable, high-throughput training and rollout across complex, asynchronous pipelines.

Qualifications

• Hands-on experience with post-training LLMs using RL (PPO/GRPO-family, RLHF, RLVR) at meaningful scale.
• Extensive distributed PyTorch training and parallelism for foundation models (FSDP, Tensor/Pipeline/Expert Parallel).
• Experience building RL environments, reward functions, verifiers, and evaluation harnesses for LLM agents.

Responsibilities

• Design, build, and scale distributed RL post-training systems across thousands of GPUs.
• Develop high-throughput rollout generation and integrate inference engines (vLLM, SGLang).
• Create sandboxed RL environments with multi-turn tool use and broad multimodal support.

Skills

Post-training LLMs with RL

Distributed PyTorch

RL environments & reward design

RL post-training frameworks

GPU clusters & NCCL/MPI

Tools

vLLM

SGLang

veRL

OpenRLHF

TRL

Ray

Job description

Speedrun Talent Network seeks an expert to design, build, and scale distributed RL post-training systems for large-scale models. You will orchestrate trainer, rollout, environment, and reward workloads across thousands of GPUs and integrate inference engines such as vLLM and SGLang.

You will develop RL environments and reward infrastructures, verifiers, and evaluation harnesses, ensuring stable, high-throughput training and rollout across complex, asynchronous pipelines.