Software Engineer, Compute and Isolation
Baseten · San Francisco
About this role
**ABOUT BASETEN** Baseten powers mission-critical inference for the world’s most dynamic AI companies (e.g., Cursor, Notion, OpenEvidence, Abridge, Clay, Gamma, Writer). By uniting applied AI research, flexible infrastructure, and seamless developer tooling, we help frontier AI companies bring cutting-edge models into production. We’re growing quickly and recently raised our **$1.5B Series F**. Join us and help build the platform engineers turn to ship AI products. --- **THE ROLE** Every agent on Baseten Sandboxes gets its own **microVM**, where it writes and runs code, calls tools, and keeps state across long tasks. Agents pause and resume constantly—so the underlying system must **start, stop, snapshot, fork, and resume fast**, while keeping every tenant isolated. The **Agent Infra** team builds this layer. Sandboxes run in microVMs with **custom Linux kernels** on **bare-metal servers** we operate. You’ll own the microVM lifecycle and performance on each host, including guest/host identity, storage attachment, and secure virtual networking devices. Baseten is also building a cloud—not just using one. The team is small, with big ownership and fast shipping. --- **RESPONSIBILITIES** - Own the **microVM lifecycle** for agent sandboxes: start, stop, suspend, resume, snapshot, and fork - Build and tune **custom guest kernels** - Tune **host CPU and memory** configuration for fast, safe sandbox execution - Manage **host-level resources** and sandbox density - Own **startup and resume performance** end to end (boot or snapshot restore → first command) - Own the **guest and host side** of each sandbox (identity, storage attachment, virtual devices) - Design **multi-tenant isolation** where the hypervisor is the security boundary - Debug virtualization and isolation issues at scale (userspace → kernel → hardware) - Contribute architecture and code back to open-source projects --- **REQUIREMENTS** - Hands-on experience with **KVM**, plus **Firecracker**, **Cloud Hypervisor**, or **QEMU** - Deep understanding of **Linux internals** (namespaces, cgroups, memory management, boot path) - Experience building **cloud infrastructure as the provider** (multi-tenant compute customers run on) - Experience building **multi-tenant infrastructure** where isolation is a security boundary - Strong systems programming background in **Rust and/or Go** (C/C++ also welcome) - Experience debugging **performance and correctness** across userspace/kernel boundaries - You move fast and like owning 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 (containerd, runc, gVisor, Kata Containers) - Experience with **VM snapshot/restore**, fast VM resume, or custom Linux kernel tuning - Experience running virtualization on **bare metal** and tuning for specific CPU/memory architectures - Experience building sandbox/serverless/function runtime products - Familiarity with Linux performance tooling (eBPF, perf, ftrace) --- **BENEFITS** - Competitive compensation, including meaningful equity - (U.S. only) 100% coverage of medical, dental, and vision insurance for employees and dependents - Flexible PTO, including company-wide Winter Break (offices closed **Dec 24–Jan 1**) - Paid parental leave - Fertility and family-building stipend (Carrot) - (U.S. only) Company-facilitated 401(k) -
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.