Agent Infrastructure Engineer (AI)

Location: Hybrid in San Francisco or New York, United States

Salary: $165K–$330K annually, plus equity

Company

Baseten builds AI inference infrastructure, developer tooling, and cloud infrastructure for companies deploying AI models in production.

What you will do

• Own the microVM lifecycle for agent sandboxes, including startup, shutdown, suspension, resumption, snapshotting, and forking.
• Build and tune custom guest Linux kernels and optimize host CPU, memory, resource allocation, and sandbox density.
• Improve startup and resume performance from VM boot or snapshot restoration through the first command.
• Own guest- and host-side sandbox identity, storage attachment, and secure virtual networking devices.
• Design multi-tenant isolation with the hypervisor as the security boundary and debug issues across userspace, kernels, and hardware.
• Contribute architecture decisions and code to the underlying open-source projects.

Requirements

• Hands-on experience with KVM and at least one of Firecracker, Cloud Hypervisor, or QEMU.
• Deep understanding of Linux internals, including namespaces, cgroups, memory management, and the boot path.
• Experience building cloud infrastructure as a provider, including multi-tenant compute that customers run on.
• Experience building multi-tenant infrastructure where isolation is a security boundary.
• Strong systems programming experience in Rust and/or Go; C/C++ is also welcome.
• Experience debugging performance and correctness issues across userspace and kernel boundaries.
• Ability to move quickly and own problems end to end on a small team.

Nice to have

• Contributions to KVM, Firecracker, Cloud Hypervisor, QEMU, rust-vmm, or the Linux kernel.
• Experience with Xen.
• Working knowledge of container runtimes and sandboxing, including containerd, runc, gVisor, or Kata Containers.
• Experience with VM snapshot and restore, fast VM resume, or building and tuning custom Linux kernels.
• Experience running virtualization on bare metal and tuning for specific CPU and memory architectures.
• Experience building a sandbox, serverless, or function runtime product.
• Familiarity with Linux performance tools such as eBPF, perf, and ftrace.

Culture & Benefits

• Meaningful equity and competitive compensation.
• U.S. employees and dependents receive 100% coverage for medical, dental, and vision insurance.
• Flexible PTO, including a company-wide Winter Break from Christmas Eve through New Year's Day.
• Paid parental leave and a fertility and family-building stipend through Carrot.
• U.S. employees have access to a company-facilitated 401(k).
• Exposure to a variety of ML startups and related learning and networking opportunities.