AI Infrastructure
New York, NY
On-site · Full-time
$225k – $320k
annual package
New York, NY · Full-time · On-site
AI Infrastructure
On-site
New York, NY
$230k – $330k
annual package
Our client is an acclaimed serverless AI and data cloud headquartered in New York City, revered by ML engineers and frontier agent builders for eliminating YAML, Dockerfiles, and multi-minute GPU boot times. Behind their effortless Python interface sits a ground-up custom Rust hypervisor, container runtime, and memory-snapshotting stack built from first principles to boot isolated GPU containers and untrusted code sandboxes in hundreds of milliseconds.
They have retained Talent247 to appoint a Senior Rust Systems Engineer for their Serverless GPU Runtime onsite in New York, NY. Working in a compact, deeply technical systems engineering culture, the successful candidate will push Linux kernel, hypervisor, and CUDA driver virtualization boundaries to make 10,000-container GPU scaleouts feel instantaneous.
io_uring filesystem clients and zero-copy page-cache layers streaming only the exact Python bytecode and shared-library pages touched during cold start.eBPF, nftables, WireGuard, custom HTTP/gRPC multiplexers) across global GPU bare-metal and cloud fleets.nvidia-uvm, CUDA Driver API, MIG/MPS, NVLink partitioning) to eliminate driver initialization bottlenecks.Stack: Rust (async Tokio, nix/libc, io_uring), C, Linux kernel internals (KVM, cgroups v2, FUSE, userfaultfd, eBPF), CUDA Driver API, bare-metal & multi-cloud H100/B200 GPU hosts. Workplace: Onsite in New York, NY (SoHo/Downtown). Compensation: $230,000 – $330,000 base salary plus premier high-upside equity, comprehensive healthcare, daily catered lunches, and relocation support.
cuda-checkpoint), or GPU virtualization.io_uring high-throughput filesystems or eBPF packet/syscall filtering.youki, containerd, runc).