Skip to content

Software Engineer, Systems

E2BSan FranciscoOn-Site

Estimated salary, not published by the employer: Est. $138k–208k/yr

Needs a free Superjobs account. You'll finish on E2B's Ashby page.

Skill score

Summary

This engineer builds and owns the Go service on each host that creates, pauses, resumes and tears down Firecracker sandboxes, along with the host Linux networking, memory and isolation beneath it. The role suits someone with deep Linux systems experience who wants to work on fast, reliable, isolated sandbox infrastructure in person in San Francisco.

Function
Development
Estimated salary
Est. $138k–208k/yr
Estimated for a mid-level role, ~5 yrs experience, United States. Not published by the employer.
Work mode and location
On-Site, San Francisco
Experience
5+ yrs
Employment type
Full-time

Job description

Overview

E2B has started billions of Firecracker sandboxes across regions on Google Cloud, AWS and Azure. Every sandbox is a real virtual machine that has to start in well under a second, resume from a snapshot on a different host, and share a host with up to about 200 others without noticing them.

Your job will be to build and own the Go service on every host that creates, pauses, resumes and tears down sandboxes, along with the Linux underneath it: host networking, memory and isolation. You'll work closely with our sandbox team in Prague, who own the Firecracker runtime, the guest kernel, sandbox storage and live migration, and with our infrastructure engineers, who put the platform together.

We're hiring more than one engineer into this role, each with a different focus: some go deepest on Linux and how sandboxes run on the host, others on sandbox networking. Our runtime is open source at http://github.com/e2b-dev/runtime, and the sandbox package is a good place to see what you'd work on.

Roles and responsibilities

  • Making sandboxes start fast and keeping them fast as the fleet grows to millions and later billions of sandboxes, with starts under 200ms
  • Evolving the architecture of the software that runs sandboxes on every host, so it can change without disrupting them
  • Making the sandbox lifecycle reliable under failure, so no sandbox is leaked or lost when a process or host dies
  • Keeping the host OS and kernel current across the fleet, rolling out upgrades without disrupting running sandboxes
  • Building the sandbox network: isolation, egress control and addressing for untrusted workloads
  • Enforcing resource isolation between sandboxes on shared hosts, so every sandbox gets the CPU, memory and I/O it was allocated
  • Building host-level observability, so we understand what every sandbox costs the host it runs on
  • Being on call for the platform you build, alongside the rest of the platform team

This role requires a high amount of ownership. You'll write the code and the design docs for the hardest changes in your area, and ship every day.

If you aren't afraid of going deep into the Linux kernel when a problem needs it, and words like eBPF, cgroups, network namespaces, nftables, conntrack, or noisy neighbor problem sound exciting to you, we want to hear from you!

Skills required

Top skills

  • Go
  • Linux internals
  • cgroups and namespaces
  • Linux networking

All skills (8)

  • Crash recovery design
  • Performance profiling
  • Rust or C/C++
  • eBPF
  • nftables
  • Firecracker
  • Kubernetes
  • seccomp

Qualifications

  • Experience building and operating systems software across a large production fleet: node agents, container or VM runtimes, schedulers, or similar
  • Strong Go, or strong Rust or C/C++ and the willingness to pick up Go quickly; our services are in Go
  • Deep Linux knowledge: processes, CPU scheduling, memory management, cgroups and namespaces, and the network stack
  • Experience making software correct under failure: crash recovery, reconciling recorded state with what is actually running, and cleanup when a process or host dies
  • For a networking focus: Linux networking for multi-tenant workloads, such as network namespaces, virtual interfaces, iptables/nftables and egress control
  • A performance mindset: you've shaved milliseconds off hot paths, profiled production systems under load, and treat p99 as a number you own
  • Excited to work in person from San Francisco
  • Comfortable with our code and infrastructure being public

We expect this experience will often come from around five or more years of relevant work, but the depth of what you've built matters more than a specific number.

Nice to have

  • Experience with VM or container runtimes such as Firecracker, KVM, gVisor or runc, including memory management and snapshot and restore
  • GPU or accelerator infrastructure, passthrough or GPU-as-a-service
  • Kernel tracing and eBPF tooling
  • OS-level hardening, such as seccomp and Linux security modules
  • Running workloads on Kubernetes, which our fleet runs on
  • Contributions to the Linux kernel, Firecracker, CNCF or similar open source projects

Other details

We offer healthcare, vision, and dental insurance, unlimited PTO, 401k, and a variety of perks for in-office employees.

Company info

E2B is an open-source sandbox platform that gives AI agents isolated machines in which to run their code. Its sandboxes are used by developers and teams building AI agents.

More jobs at E2B