Modal
Member of Technical Staff - Inference Runtime
$220K–$300K
San Francisco · On-site · Staff · Full-time · Posted Oct 5, 2026
About the job
Build and optimize Modal’s container runtime infrastructure for fast, efficient, and securely isolated AI inference and training workloads.
What you'll do
- Improve container startup, checkpointing, restoring, snapshotting, and filesystem performance.
- Develop and optimize runtime support for GPUs, RDMA, storage, and low-copy data paths.
- Diagnose Linux and performance issues, deploy runtime and kernel changes, and contribute to gVisor where appropriate.
What they ask for
- Strong Linux systems knowledge, including processes, virtual memory, filesystems, system calls, scheduling, namespaces, cgroups, and signals.
- Experience building or debugging container runtimes, sandboxes, the Linux kernel, or similar low-level infrastructure.
- Strong programming ability in Rust, Go, or another systems language, and interest in becoming productive in both Rust and Go.
- Experience profiling and improving systems where memory movement, I/O, synchronization, or kernel interactions dominate performance.
Nice to have
- Experience with gVisor, runsc, runc, OCI runtimes, seccomp, or checkpoint/restore systems.
- Experience with Linux kernel development, GPUs, RDMA, FUSE, EROFS, direct I/O, or large-memory and multi-GPU systems.
Pay
$220K – $300K USD base salary
Tools
Rust, Go, Linux, gVisor, NVIDIA, AMD, RDMA, EROFS
AI focus
Build runtime infrastructure that accelerates and supports AI inference and training workloads, including multi-GPU and accelerator-aware snapshotting.
Model-written summary · the original posting is authoritative