Staff Software Engineer, Storage

CoreWeave•Sunnyvale, CA

2d•$188,000 - $275,000•Onsite

About The Position

CoreWeave's Storage team builds the file, block and object storage that the world's largest AI training and inference workloads run on. As a Staff Software Engineer you will own whole storage subsystems end to end, from the data path to the multi-tenant control plane, on distributed systems built in Go, Rust and C++ and delivered through Kubernetes. Storage at CoreWeave sits next to the GPUs, so your work directly determines how fast the largest AI clusters in the world train. You will set technical direction for foundational storage services built from scratch: high-performance file storage, durable block volumes and the control plane that governs tenancy, provisioning and data protection at exabyte scale. Own the architecture and delivery of multi-tenant file and block storage services, data path through control plane, running on CoreWeave's own hardware fleet. Design for the IO patterns AI pipelines create: small-file and metadata-heavy workloads, IOPS at scale and the sustained throughput that keeps GPUs fed. Ship storage as a first-class Kubernetes citizen: CSI drivers, dynamic provisioning, snapshots, resize and RWO block volumes backed by public SLOs. Work with technologies such as RDMA, GPU Direct Storage, RoCE, InfiniBand, SPDK and distributed filesystems to raise storage performance and efficiency. Drive reliability, durability and observability across the storage stack using telemetry, metrics and dashboards. Set the technical bar: lead design reviews, mentor senior engineers and shape how storage is built across the organization.

Requirements

Bachelor's or Master's degree in Computer Science, Engineering or a related field.
10+ years of experience in storage systems, database internals or infrastructure engineering.
A track record of building storage or database systems others rely on: a founding or lead engineer on a shipped storage product, published systems work, or equivalent depth.
Strong production experience with distributed file systems, block storage or storage engines, including protocols such as NFS or S3 and systems such as Ceph, Lustre or similar.
Depth in the storage data path: IO performance, caching, crash consistency and resilience under load.
Expert in a systems programming language: Rust, C++ or Go.
Experience building on Kubernetes: controllers, operators, CRDs and CSI.
Strong debugging and problem-solving skills in distributed, high-performance environments.
Clear communicator, able to work collaboratively across teams and share technical insights effectively

Nice To Haves

You love owning a system from whiteboard to production
You're curious about what storage looks like when it lives next to the GPUs
You're an expert in making distributed systems fast, durable and simple to operate

Responsibilities

Own the architecture and delivery of multi-tenant file and block storage services, data path through control plane, running on CoreWeave's own hardware fleet.
Design for the IO patterns AI pipelines create: small-file and metadata-heavy workloads, IOPS at scale and the sustained throughput that keeps GPUs fed.
Ship storage as a first-class Kubernetes citizen: CSI drivers, dynamic provisioning, snapshots, resize and RWO block volumes backed by public SLOs.
Work with technologies such as RDMA, GPU Direct Storage, RoCE, InfiniBand, SPDK and distributed filesystems to raise storage performance and efficiency.
Drive reliability, durability and observability across the storage stack using telemetry, metrics and dashboards.
Set the technical bar: lead design reviews, mentor senior engineers and shape how storage is built across the organization.