1 day ago
Base Salary
$250k - $350k/yr
Responsibilities
- Build and optimize the host software that starts and manages sandboxes at fleet scale, targeting startup times under 200 milliseconds.
- Evolve sandbox-host architecture so it can change without disrupting running sandboxes.
- Implement reliable sandbox lifecycle handling, including crash recovery, reconciliation, and cleanup after process or host failures.
- Keep host operating systems and kernels current through non-disruptive fleet-wide upgrades.
- Build isolated sandbox networking with addressing and egress controls for untrusted workloads.
- Enforce CPU, memory, and I/O isolation between sandboxes sharing a host.
- Build host-level observability for understanding sandbox resource costs.
- Write code and design documentation for complex systems changes and participate in on-call operations.
Requirements
- Experience building and operating systems software across a large production fleet, such as node agents, runtimes, or schedulers.
- Strong Go skills, or strong Rust or C/C++ skills with willingness to learn Go quickly.
- Deep Linux knowledge covering processes, CPU scheduling, memory management, cgroups, namespaces, and networking.
- Experience making software correct under failure, including crash recovery, state reconciliation, and cleanup after process or host failures.
- For networking-focused work, experience with Linux networking for multi-tenant workloads, virtual interfaces, iptables or nftables, and egress control.
- Experience profiling production systems under load and optimizing performance-sensitive paths.
- Approximately five or more years of relevant experience is expected, although depth of experience is prioritized over a specific number.
- Ability and interest to work in person from San Francisco.
- Bonus qualifications include experience with Firecracker, KVM, gVisor, runc, GPU or accelerator infrastructure, eBPF tooling, seccomp, Linux security modules, Kubernetes, or contributions to relevant open-source projects.
Benefits
- In-person work from the San Francisco or Prague office, with four days onsite and one day working from home.
- Healthcare, vision, and dental insurance.
- Unlimited paid time off.
- 401(k) plan.
- Additional perks for in-office employees.
About E2B
E2B builds the Agent Cloud—secure, scalable sandboxes and infrastructure for interpreting and running AI‑generated code—delivered via APIs to developers and AI labs building agents. The privately held, Series A company is headquartered in San Francisco and is used by teams behind agents at Groq, ClickUp, and Lindy; its team includes original authors of Firecracker, the AWS microVM technology behind Lambda.
