Software Engineer (Networking, Agent Infr)
Posted Updated
Description
- The Agent Infra team builds its own networking stack for sandboxes, designed from the start for that scale
- It runs in user space, with VPP (FD.io) as its main data plane technology, and is built from virtualized network functions (VNF/NFV): carrier-grade, low-latency packet processing designed for very large numbers of agents
- You’ll build multi-tenant virtual private networking that connects each customer’s agents and services privately and securely, using modern overlay technologies
- We are building a cloud, not just using one. We want engineers who expose the network as a programmable API and write the code, not operators who push router and switch configuration
- The team is small and ownership is big: we ship in weeks and months, not multi-year programs
- This role is separate from Baseten’s GPU-to-GPU networking. Here you’ll own how sandboxes connect to each other, to Baseten services, to customer networks, and to the internet
- If you’ve built user-space data planes, virtualized network functions, or multi-tenant cloud networking and want to design one for agents, we’d love to talk
- Design and build the user-space, VPP-based data plane that connects microVM sandboxes to each other, to Baseten services, and to the internet
- Build multi-tenant virtual private networking and overlays (for example EVPN, GENEVE, VXLAN) across regions
- Expose the network as a programmable API that the rest of the platform calls
- Cut latency and raise connection scale for agent-to-agent and agent-to-service traffic
- Build routing that sends each request to the nearest region and keeps connections working as sandboxes suspend and resume
- Build secure ingress and egress for sandboxes, including firewalling, static egress IPs, and custom domains with managed TLS
- Build L7 proxies for request inspection, filtering, secret injection, and auditing
- Own per-sandbox addressing, including IPv6, and network isolation between tenants
- Build observability for the packet path across guest, host, and network, and debug production issues end to end
- Work closely with the compute, storage, GPU infrastructure, and datacenter teams
Benefits
- Remote-first work environment. The Baseten team is welcome to work from wherever they want; fully remote, in our San Francisco office, or a mix of both. Today, our team (including our founding team) is spread across the United States, Canada, and Armenia. We provide a $1,000 stipend for you to make your home-office comfortable and productive
- Regular in-person team summits. We get together as a team three times a year to plan, workshop, and most importantly, get to know each other better
- Unlimited PTO. We ask that everyone take at least 4 weeks of vacation. And we have a company-wide break between Christmas and New Year’s Day
- Full healthcare coverage. Medical, dental and vision insurance for you and your family
- Paid parental leave. 16-weeks fully paid parental leave (adoptive and non-birth parents included) and flexibility with schedules while returning to work
- Company-sponsored 401(k) for you to contribute to
- Learning and development budget. We encourage you to take classes, attend conferences, and invest in your craft and we’ll cover expenses to make it happen
- Experience building multi-tenant networking in production (VXLAN, GENEVE, VRF)
- You write production code: strong systems programming in Rust, Go, and/or C/C++. Network architects are welcome if they build, not only configure
- You move fast and like owning a problem end to end on a small team
- Experience debugging latency and throughput problems with packet captures, tracing, and profiling
- Experience building a software-defined network, overlay network, or packet-processing data plane in production
- Experience building cloud networking as the provider, not only consuming a cloud provider’s network
- Solid grounding in TCP/IP and routing
- Experience with user-space packet processing (DPDK, VPP, or similar) or virtualized network functions (VNF/NFV)
- Experience building or operating proxies, load balancers, or L4/L7 gateways at scale
- Experience with VPP / FD.io, our main networking technology
- Experience with GENEVE or VXLAN overlays and EVPN
- Linux kernel networking: eBPF/XDP, netfilter or nftables, network namespaces
- Familiarity with virtual machine networking: virtio-net, vhost, TAP devices, or networking for Firecracker or Cloud Hypervisor
- Experience running IPv6 at scale
- Background in telecom user-plane or 5G core networking, or in low-latency trading networks
- Experience with multi-region traffic routing, anycast, or BGP
- Experience with TLS termination, DNS, and egress security controls
Skills
Sign in to read and join the discussion.