Daytona
Senior Software Engineer, GPU Sandboxes
San Francisco · Senior · Full-time · Posted Oct 30, 2026
- Experience
- 5+ years of systems engineering in production environments.
About the job
Build and operate an isolated GPU sandbox runtime and the fleet systems that make GPU-backed workloads available through an API.
What you'll do
- Build and harden the GPU sandbox runtime, including virtualization, passthrough, and guest driver/CUDA lifecycle.
- Reduce cold-start latency and implement pause, snapshot, and fork semantics for GPU workloads.
- Build GPU-aware scheduling, monitoring, failure detection, and remediation; manage security boundaries and debug hardware and driver issues.
What they ask for
- 5+ years of systems engineering in production environments.
- Hands-on experience with Linux virtualization internals: KVM, QEMU, libvirt, VFIO, and IOMMU.
- Strong Go, Rust, or C/C++ skills and Linux kernel-adjacent debugging skills.
- Experience operating GPUs in production.
Nice to have
- Contributions to QEMU, KVM, Firecracker, Cloud Hypervisor, or similar.
- NVML/DCGM tooling, GPU telemetry, driver-level work, RDMA, GPUDirect, or multi-GPU interconnect experience.
Tools
KVM, QEMU, VFIO, IOMMU, Go, Rust, C/C++, CUDA
AI focus
Build GPU infrastructure for AI agent workloads, reinforcement learning, and developer environments.
Model-written summary · the original posting is authoritative