Senior Software Engineer II, Applied Training
$182k - $242k • Remote • Sunnyvale, CA / Bellevue, WA
Posted 5d ago
About the job
CoreWeave is seeking an early member for its Applied Training team to build the research infrastructure that enables AI pioneers to train models with confidence. This role addresses the problem of AI labs spending valuable research time on cluster setup and operations instead of innovation. You will contribute to the roadmap and develop solutions for our Kubernetes-native research cluster platform or sandbox client for agentic training and evaluation, aiming to provide frontier lab-level infrastructure to all CoreWeave customers.
Responsibilities
- Contribute to the roadmap for Applied Training, identifying key features for new workloads.
- Collaborate closely with customers and internal teams building cloud-native primitives.
- Design and build a complete research cluster experience, including CLI, job configuration, Kubernetes operators, and daemons, to solve researcher challenges.
- Own the Python SDK for sandbox infrastructure, enabling large-scale agent rollouts and benchmarks.
- Write documentation for popular OSS training frameworks on CoreWeave.
- Understand and incorporate the needs of large AI labs running thousands of GPUs into system designs.
Requirements
- 8-12+ years of experience building distributed systems, ML infrastructure, or developer platforms.
- Deep Kubernetes experience, including custom controllers, operators, scheduling, CRDs, and large-scale workload orchestration.
- Understanding of researcher productivity needs, including code distribution and fast iteration cycles.
- Familiarity with distributed job scheduling, rank initialization, and scaling challenges in training.
- Proven experience shipping production infrastructure systems.
- Strong communication skills to translate researcher needs into system designs.