Large Language Model Engineer for RL Framework and RL Inference
Active AI foundation-model company offering multimodal models, consumer AI products, and an enterprise/developer API platform.
Funding history
About MiniMax Group Inc.
MiniMax develops proprietary multimodal foundation models spanning language, video, speech, and music, and operates AI-native products including MiniMax Agent, MiniMax Design, MiniMax Audio, Talkie, Hailuo AI, and an Open Platform.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will build and optimize an RL framework and rollout systems for large-scale RL training. You will improve scheduling, latency, KV cache management, throughput, and inference serving; support complex interactions and tool calls; and co-design RL algorithms and inference systems with algorithm engineers.
Requirements
- Strong programming and system-design skills
- Deep understanding of large language model inference systems, RL engineering, distributed systems, or GPU performance optimization
- Ability to identify system bottlenecks in RL training or rollout pipelines and deliver solutions
- Knowledge of RL, inference systems, and agentic workflows
- Experience with SGLang, vLLM, TensorRT-LLM, PyTorch, or Megatron-LM is preferred
- Experience with KV cache management, continuous batching, low-precision quantization, distributed serving, or communication optimization is preferred
Responsibilities
- Build and optimize the Forge RL framework for large-scale RL training
- Optimize RL rollout systems for multi-turn interactions, environment calls, and tool calls
- Improve scheduling, tail latency, KV cache management, and throughput
- Build high-performance inference serving for RL trajectory generation
- Co-design RL algorithms and inference systems with algorithm teams