Principal Systems Software Engineer

Principal Systems Software Engineer Crusoe's mission is to accelerate the abundance of energy and intelligence. We’re crafting the engine that powers a world where people can create ambitiously with AI — without sacrificing scale, speed, or sustainability. Be a part of the AI revolution with sustainable technology at Crusoe. Here, you'll drive meaningful innovation, make a tangible impact, and join a team that’s setting the pace for responsible, transformative cloud infrastructure. PRINCIPAL SYSTEMS SOFTWARE ENGINEER San Francisco, Sunnyvale (On-site) ABOUT THE ROLE Crusoe is seeking a Principal Systems Architect to serve as the visionary lead for our next-generation AI infrastructure. This is a role for an industry-recognized expert who has already "seen the movie" at hyperscale and is ready to redefine the I/O path for the age of generative AI. In this position, you aren't just building a cloud; you are designing the fluid fabric that unifies Bare-Metal-as-a-Service (BMaaS), Intelligent IaaS, and Elastic CaaS into a single, high-performance pool of intelligence. You will bridge the gap between silicon and software, advising executive leadership on critical hardware/software co-design pivots while remaining hands-on enough to lead R&D teams in shipping production-grade kernel and orchestration code. WHAT YOU’LL BE WORKING ON UNIFYING PILLARS: - Bare-Metal-as-a-Service (BMaaS): Architecting the systems that deliver raw GPU throughput via zero-latency InfiniBand/RDMA fabrics, ensuring that massive-scale training workloads perform at the theoretical limits of the hardware. - Intelligent IaaS: Designing a highly optimized, thin virtualization layer using KVM or custom micro-VMs. Your goal is to provide enterprise-grade isolation and multi-tenancy without the virtualization tax on performance. - Elastic CaaS: Building a high-performance container substrate (utilizing Kubernetes or Slurm) that allows AI workloads to burst and scale across heterogeneous GPU nodes instantly. MASTERING THE I/O PATH - Leading the architectural design of our internal cloud fabric, drawing on experience building at hyperscalers (OCI, AWS, GCP) or top-tier neoclouds. - Driving the technical roadmap for SR-IOV, RDMA, and virtualized GPU scheduling. You will hold the expertise (and potentially the patents) that define how we handle high-density GPU interconnects. - Leading elite R&D workstreams to prototype and productionize new ways of managing memory, networking, and compute that don't yet exist in standard cloud distributions. A DAY IN THE LIFE - Drafting white papers and RFCs that define the next two years of Crusoe’s compute and networking stack. - Working alongside Staff and Senior engineers to debug complex race conditions in the I/O path or optimize kernel-level memory pinning for GPU clusters. - Representing Crusoe in open-source communities or industry forums to influence the direction of cloud-native AI infrastructure. YOU WILL THRIVE IN THIS ROLE IF YOU HAVE: - 12+ years of experience designing and shipping core infrastructure at a major hyperscaler (OCI, AWS, Azure, GCP) or a specialized HPC cloud. - Deep, authoritative knowledge of the Linux kernel, virtualization internals (KVM/QEMU/Firecracker), and high-performance networking (RoCE v2, InfiniBand). - Proven ability to design software that maximizes the performance of NVIDIA/AMD GPUs and high-speed NICs. - Experience leading cross-functional R&D teams through high-ambiguity projects and delivering production-ready systems. - A portfolio of significant contributions to the field, which may include patents, major open-source contributions, or published research in distributed systems. - The rare ability to explain the nuances of memory-mapped I/O to an engineer and the business value of a new fabric architecture to a Board member. BENEFITS - Industry-leading competitive pay and significant equity - Restricted Stock Units in a fast-growing, well-funded technology c ... (truncated, view full listing at source)