R
Posted 11 hours ago
Member of Technical Staff — Training
RadixArkMember of Technical Staff — Training
Requirements
3+ years ML systems experience, Large-scale training infrastructure experience, Agentic post-training systems experience, Training/inference correctness experience, Performance debugging experience
Skills
Machine LearningLLMDistributed Training
About the role
Responsibilities
- Contribute to open-source large-scale post-training infrastructure Miles, and inference system SGLang
- Optimize throughput, scalability, and hardware efficiency
- Improve reliability and fault tolerance for long-running training jobs
- Develop training frameworks and infrastructure tooling
- Collaborate with model researchers to support frontier experiments
- Debug and resolve cross-layer performance bottlenecks
- Build observability systems for training performance and reliability
- Drive capacity planning and cluster utilization strategies
- Contribute to long-term training infrastructure architecture
Requirements
- 3+ years of experience in ML systems or large-scale training infrastructure
- Experience building or operating large-scale agentic post-training systems
- Experience working on training/inference correctness or precision-related problems
- Experience debugging performance and stability issues in large post-training jobs
- Experience improving training or inference efficiency
Preferred Qualifications
- Experience training 100+ billion-parameter models
- Experience with train/inference optimization for large-scale RL or other production workloads
- Familiarity with training stacks (e.g. Megatron-LM, FSDP, torchtitan) and inference stacks (e.g. SGLang, vLLM)
- Familiarity with post-training frameworks (e.g. Miles, Slime, veRL, Prime-RL, AReaL)
- Experience with RDMA, InfiniBand, NVLink, NCCL/RCCL, or high-speed GPU interconnects
- Contributions to ML systems open-source projects
- Experience with checkpointing, fault recovery, and elastic training
- Experience building infrastructure for agentic post-training, such as async rollout pipelines, sandbox, or harness systems
About the Company
RadixArk is an infrastructure-first company built by engineers who've shipped production AI systems, created SGLang, and developed Miles. We are on a mission to democratize frontier-level AI infrastructure by building world-class open systems for inference and training.
ScoutJobs Agent
Get matches like this delivered daily
Sign up free — we'll pull jobs that fit your CV from across the web and rank them for you.
Get started — it's freeMember of Technical Staff — Training
RadixArk · Palo Alto
