Staff Software Engineer (Storage Control Plane)

About this role

• We’re looking for a Staff Storage Engineer to play a key role in designing, building, and operating the control plane for our high-performance AI storage platform
• You’ll help evolve CoreWeave’s storage systems by building reliable, scalable, and high-throughput solutions that power some of the largest and innovative AI workloads in the world
• This role involves close collaboration with teams across infrastructure, compute, and platform to ensure our storage services scale automatically and seamlessly while maximizing performance and reliability
• Design and implement a highly scalable multi-tenant control plane that supports CoreWeave’s growing AI storage and cloud infrastructure needs
• Contribute to the development of exabyte-scale, S3-compatible object storage, distributed file system and integrate dedicated storage clusters into diverse customer environments
• Work with technologies such as RDMA, GPU Direct Storage, RoCE, InfiniBand, SPDK, and distributed filesystems to optimize storage performance and efficiency
• Participate in efforts to improve the reliability, durability, and observability of our storage stack
• Collaborate with operations teams to monitor, analyze, and optimize storage systems using telemetry, metrics, and dashboards to improve performance, latency, and resilience
• Work cross-functionally with platform, product, and infrastructure teams to deliver seamless storage capabilities across the stack
• Share your knowledge and mentor other engineers on best practices in building distributed, high-performance systems- Experience with one or more storage protocols (e.g. S3, NFS) and file systems such as Ceph, DAOS, or similar
• Bachelor’s or Master’s degree in Computer Science, Engineering, or a related field
• Solid understanding of cloud-native infrastructure, Kubernetes, and scalable system architecture
• Proficiency in a systems programming language such as Go, C++, or Rust
• Strong hands-on experience with object storage or distributed f...