Site Reliability Engineer

Anduril Industries is a defense technology company with a mission to transform U.S. and allied military capabilities with advanced technology.

The Imaging team builds and fields state-of-the-art camera and sensor systems deployed to solve real security challenges for the United States and its allies.

We're looking for a Site Reliability Engineer to join the Imaging team. This is not a product development role, and it isn't a traditional cloud-SRE role either. You are the frontline for keeping fielded imaging systems alive: the person field personnel and customer-support escalations turn to when a deployed system isn't behaving.

Responsibilities

• Own fielded system reliability. You are responsible for the health and uptime of deployed imaging systems. When issues arise--whether the root cause is in the network, the calibration, an upgrade, or the sensor hardware itself--you triage, diagnose, and drive resolution.
• Run point on escalations. You'll be first and second line of response for issues coming through our support channels and Anduril's customer-support pipeline, acting as the deep-expertise backstop the rest of the funnel escalates to.
• Turn fires into runbooks. Recurring issues shouldn't be solved live twice. You'll build and maintain runbooks, diagnostics, and self-service tooling that reduce repeat problems and shrink the support load over time.
• Hold the boundary with engineering. You own everything short of a code fix. When an issue turns out to be a genuine software defect, you'll cleanly reproduce it, document it, and hand it off to the Mission Software Engineers--protecting their focus by keeping frontline support where it belongs.
• Feed reliability back into the product. You'll turn what you learn in the field into signals that make our systems more supportable--better observability, safer upgrades, more graceful failure.
• Travel expected approximately 20% of the time for field support and deployment windows.

Requirements

• 3+ years in SRE, DevOps, field/systems engineering, or production support of deployed hardware/software systems--with real ownership after systems ship, not just standing them up
• Strong Linux fundamentals, including comfort troubleshooting real networking issues (IP, routing, VPNs, connectivity in constrained or field environments)
• Demonstrated ability to diagnose and resolve issues across system boundaries (networking, services, hardware interaction) without always having full visibility into every component
• Comfortable owning a structured on-call rotation, including scheduled after-hours and weekend coverage
• Strong written and verbal communication skills, including the ability to run a remote troubleshooting session with a non-technical operator and document what happened
• Eligibility to obtain and maintain a U.S. Secret clearance

Preferred Qualifications

• Experience supporting fielded or deployed systems, not just development environments
• Experience with fielded hardware or sensor systems (EO/IR, optical, or similar), including familiarity with sensor calibration
• Scripting for diagnostics and automation (Python, Bash, or similar)
• Familiarity with Nix or NixOS--uncommon, but valuable on our stack
• Familiarity with systemd service management and observability practices on Linux
• Familiarity with incident tooling (PagerDuty or equivalent)

What Makes Someone Successful Here

The people who thrive in this role share a few traits that don't always show up on a resume:

• They close loops. When they pick up a problem they own it until it's resolved, not until it's someone else's problem.
• They stay calm under pressure. A system down in the field with an operator waiting is a high-stress moment. They work the problem methodically instead of reacting.
• They communicate proactively. They don't wait to be asked for a status update. They document what they find, share what they learn, and keep the people who need to know informed.
• They're comfortable with ambiguity at the stack boundary. When a device behaves unexpectedly in the field, the root cause could be anywhere. They have a methodology for narrowing it down systematically rather than guessing or waiting for someone else to have the answer.
• They know where the line is. They own frontline support fully, and they know when an issue is a true code-level defect that belongs with engineering--handing it off cleanly rather than either sitting on it or tossing it over the wall half-formed.