RL Systems Architect: Scalable AI Training & Infra

Bytedance; San Jose (CA); On-site; USD 244,800 - 450,000; Full time

Job summary

ByteDance is seeking a Research Engineer for RL Systems & Infrastructure in San Jose. You will design end-to-end RL systems for large-scale models, build fault-tolerant infrastructure, and optimize distributed training across GPU clusters.

Collaboration with researchers to convert ideas into robust, production-ready solutions is essential. The role offers exposure to cutting-edge RL and ML workflows within a globally diverse team and a strong emphasis on reliability, performance, and

Qualifications

• Experience with distributed systems and large-scale ML infrastructure.
• Experience building or optimizing RL/LLM/multimodal training pipelines.
• Solid Python/C++ programming and familiarity with PyTorch and distributed training.

Responsibilities

• Design end-to-end RL systems for large-scale models, covering rollout, training, evaluation, and deployment.
• Develop scalable, fault-tolerant RL infrastructure for dynamic workloads and heterogeneous compute environments.
• Optimize distributed training across GPU clusters for throughput and stability.
• Collaborate with researchers to translate ideas into production-grade implementations.
• Build tooling and observability for RL training systems.

Skills

Distributed systems

Large-scale ML systems

Deep learning infrastructure

Python

C++

PyTorch

RL training workflows

GPU optimization

Tools

Distributed training frameworks

Job description

ByteDance is seeking a Research Engineer for RL Systems & Infrastructure in San Jose. You will design end-to-end RL systems for large-scale models, build fault-tolerant infrastructure, and optimize distributed training across GPU clusters.

Collaboration with researchers to convert ideas into robust, production-ready solutions is essential. The role offers exposure to cutting-edge RL and ML workflows within a globally diverse team and a strong emphasis on reliability, performance, and