SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 215,000 - 255,000 / annual
Crusoe is a vertically integrated AI infrastructure company building the complete stack from energy generation through AI workload execution. The company is solving the power bottleneck constraining AI compute by combining energy, manufacturing, data center construction, and cloud services into a unified platform.
You will architect and own Crusoe's storage network end-to-end, from spine to leaf switches. This is a foundational technical role with full ownership of a dedicated RDMA-based storage fabric engineered for lossless, low-latency performance—distinct from the general compute network. You are not maintaining an existing system; you are designing the backbone of storage infrastructure from the ground up and setting the standard for how it operates at scale.
Key responsibilities include:
- Architect and operate the storage network end-to-end, making core design decisions that shape performance at scale
- Serve as the first and deepest line of defense for cluster fabric issues—tracing packets through switch pipelines, reading per-switch configuration, and finding root cause under production pressure
- Go beyond the switch layer into the OS network stack, NIC drivers, and software layers when needed
- Build automation and tooling that enables the network to run itself, turning debugging and monitoring into a system rather than a scramble
- Partner closely with SDN and Network Edge teams to bring software-defined thinking into traditionally hardware-focused network operations
- Act as the connective layer between network and storage domains, translating network reality into storage context and storage needs into network architecture
You will be the deep technical authority the organization leans on as the storage infrastructure scales. This role requires someone who thrives on ownership, can operate under production pressure, and can communicate complex technical realities across domains.
Requirements:
- Battle-tested L2/L3 networking experience running production switches under production pressure
- Strong scripting and debugging instincts to chase problems down to the TCP/IP level
- Sense of ownership and willingness to be the person who gets paged and solves issues end-to-end
- Communication skills to operate as a translator between technical domains
- Ability to pass a background check
Bonus qualifications:
- Spine-layer switch expertise
- Hands-on experience with RDMA, RoCEv2, DCQCN/PFC, or similar lossless, low-latency technologies
- Software-defined networking mindset with systems and automation thinking
- Familiarity with client-side network behavior (OS stack, NIC drivers)
- Exposure to storage protocols like NFS
- Background at a storage-infrastructure company or deep hands-on switch experience from hyperscale/cloud environments (AWS, Azure, GCP)
