SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 220,000 - 292,000 / annual
Anduril Industries is a defense technology company transforming U.S. and allied military capabilities through advanced technology. The Sentry Tower Software team develops robotic systems providing force protection capabilities for secure perimeters, leveraging advanced sensor fusion and autonomy to render environmental activity into Lattice's common operating picture.
You will own the platform overlay and networking infrastructure for Sentry Tower systems across Counter-Intrusion and Air Defense product lines. Every tower boots from a common NixOS-based software platform that defines base images, system services, compute configuration for each hardware generation, and network topology connecting the tower's payloads. This is a systems and infrastructure role with direct product consequences: the tower's network carries camera, radar, GPS, and sensor-board traffic over VLANs and bridges across mesh radio, cellular, and satellite backhaul on hardware at the far end of a long support loop. Field failures directly reflect platform decisions.
Key responsibilities include:
- Own the shared NixOS overlay base: system modules, compute platform definitions, common tower services, and image definitions that every product variant composes from
- Own tower network topology end-to-end: VLAN and subnet allocation, bridging, link ranking and failover across heterogeneous backhaul, NAT and firewall policy, and physical-port-to-payload mapping for each carrier board
- Define and enforce the integration contract that business line teams use to add network-attached payloads, ensuring extensions rather than divergence
- Drive consolidation of legacy per-product configurations into the common overlay with build-time checks preventing drift
- Support and extend the real-time messaging layer as services migrate from legacy pub-sub to next-generation replacement, including topic and schema contracts
- Debug problems crossing every layer: kernel and device tree, systemd unit ordering, Nix evaluation, link-layer configuration, and sensor daemons
You will be a primary steward of the build and configuration system. Images are fully declarative and reproducible, spanning several compute generations and a dozen product variants, with active consolidation of legacy configurations into a common, composable base.
REQUIREMENTS:
- 4+ years of professional software engineering with substantial time in Linux systems, embedded platforms, or infrastructure work
- Deep, practical networking knowledge: VLANs, bridging, routing and policy routing, DHCP, multicast, NAT, and firewalling; fluency with ip, tcpdump, and Wireshark for diagnosing real traffic
- Strong experience with declarative configuration management and reproducible builds; Nix or NixOS preferred; adjacent depth in Bazel, Buildroot, Yocto, or infrastructure-as-code tooling (Terraform, Ansible, Pulumi) acceptable
- Proficiency in Rust or C++, sufficient to read and modify services on top of the platform, not just configuration
- Comfort operating and troubleshooting real hardware: serial consoles, provisioning and reflashing, boot problems, and remote debugging of unreachable systems
- Demonstrated ability to define technical interfaces that other teams build against, holding the line through code review and documentation rather than authority
- US person status required with eligibility to obtain and maintain a US security clearance
NICE TO HAVE:
- Hands-on NixOS experience including writing modules and maintaining flakes
- Python for tooling and automation
- Experience with NVIDIA Jetson platforms (Xavier, Orin, Thor) including BSP integration, device trees, and JetPack upgrades
- Fleet-scale operational experience: over-the-air update systems, observability pipelines, or long-lived embedded deployments
- Familiarity with CAN, SNMP, or managed-switch configuration in embedded context
- Experience setting up and maintaining CI/CD for large build graphs and hardware-in-the-loop pipelines
