Senior Software Engineer, Sandboxes & Virtualization in Sunnyvale

Core Weave is The Essential Cloud for AI™. Built for pioneers by pioneers, Core Weave delivers a platform of technology, tools, and teams that enables innovators to build and scale AI with confidence. Trusted by leading AI labs, startups, and global enterprises, Core Weave combines superior infrastructure performance with deep technical expertise to accelerate breakthroughs and turn compute into capability. Founded in 2017, Core Weave became a publicly traded company (Nasdaq: CRWV) in March 2025.

The Runtime & GPU Systems team builds and operates secure, high-performance environments for multi-tenant Kubernetes platforms  team sits at the intersection of container runtimes, lightweight virtualization, GPU infrastructure, and Linux systems engineering. We partner closely with security, platform, and GPU infrastructure teams to deliver robust, isolated, and efficient execution environments for AI-driven workloads.

As a GPU & Runtime Systems Engineer, you will design, build, and evolve secure sandboxed runtime environments for Kubernetes workloads, focusing on runtime isolation, performance, and security. You will integrate container runtimes, lightweight VMs, and virtualization technologies to support GPU-accelerated workloads in multi-tenant environments. Day-to-day, you'll develop GPU-aware sandboxing and scheduling strategies, optimize containerization and I/O performance for latency-sensitive workloads, and influence architectural decisions across Linux internals, container runtimes, virtualization layers, and GPU drivers.

• Design and implement secure execution environments for containerized and virtualized workloads.
• Build GPU‑aware scheduling, isolation, and resource management strategies for multi‑tenant workloads.
• Optimize container, VM, and I/O performance across GPU‑accelerated workloads.
• Conduct profiling, benchmarking, and performance tuning for runtime, virtualization, and GPU stacks.
• Contribute to architectural decisions across Linux internals, container runtimes, virtualization layers, and GPU drivers.
• Collaborate with security, platform, and infrastructure teams to define and implement runtime isolation and performance standards.

• 3+ years of experience in systems, platform, infrastructure, or production engineering at scale.
• Strong hands‑on experience with Kubernetes, container orchestration, and cloud architectures, including controllers, operators, or scheduling extensions.
• Experience designing, implementing, or operating secure execution environments (container runtimes, sandboxed workloads, or virtualized systems).
• Practical experience with lightweight virtualization and sandboxing technologies (e.g., Kata Containers, gVisor, Kube Virt, QEMU).
• Experience supporting GPU‑accelerated workloads in multi‑tenant environments, including GPU scheduling, isolation, device passthrough, mediated devices, or virtualization.
• Proficient in systems‑oriented programming (Go, C/C++, Rust, Bash) with strong Linux internals knowledge.
• Skilled at diagnosing and resolving complex performance, reliability, or isolation issues across containers, VMs, and infrastructure.
• Experienced in profiling, benchmarking, and tuning performance across runtime, virtualization, and GPU stacks.

• Experience building systems for safely executing untrusted or sensitive workloads in shared environments.
• Familiarity with GPU drivers and low‑level virtualization or I/O optimization techniques.
• Experience defining threat models and implementing runtime security policies in multi‑tenant systems.

We believe in investing in our people, and value candidates who can bring their own diversified experiences to our teams – even if you’re not a 100% skill or experience match. If some of this describes you, we’d love to talk.

• You love building high‑performance systems that operate reliably under extreme scale and demand.
• You’re curious about the intersection of security, virtualization, Kubernetes, and GPU infrastructure.
• You’re an expert in reasoning about trade‑offs between isolation, performance, and operability.