Responsibilities

We're hiring one of the first software engineers on our electro-optical team. Today that means long-range EO/IR sensors. In future it could mean sensors on mobile and airborne platforms. Wherever the sensor sits, the job is the software that makes it, and every sensor around it, act as one system.
This is a software engineering role first. You don't need to be a camera or computer vision specialist. You do need to be an excellent engineer who is curious about physical systems and understands how real sensors behave: they drift, they disagree with each other, and a field test doesn't care what the simulation said.
You'll join a small, elite team that takes its own work into the field. You'll test live, change code live, and see how it holds up on the next flight. The capabilities you build will push the frontier of what these systems can do, and few engineers anywhere get to work this close to the hardware, the mission and the result.
We're also changing how we build software. The aim is to take a feature or a new capability and develop as much of it as possible without real hardware, then push it up the chain: through simulation, through hardware in the loop, and on to full end-to-end tests in the real world. Simulation will keep improving, but real hardware will always have the final say. So we're building the harnesses and graphs that carry a capability through every one of those stages, with AI doing as much of the work as it can be trusted with. The end state is fully autonomous, end-to-end tests run against real hardware at our test sites. This is what you'll be building.
What You'll Build:
Software that works on any sensor. Code that stays generic across many different camera heads and configurations while still hitting its performance bar, and flexes to whatever the mission and the customer demand.
The path from desk to field. Harnesses and graphs that take a capability from first commit through simulation and hardware in the loop to a full end-to-end test, with as little manual effort as possible.
Autonomous field testing. The system that runs complete tests against real hardware at our test sites on its own, and reports back what happened.
Better simulation and replay. So more of every feature can be built and proven before it touches hardware.
Multi-sensor integration. The interfaces, timing and data flow that let radar, cameras and other sensors work as one system.
Fielded software. Deployment, monitoring and live fixes for systems running on edge hardware at test and customer sites.
This Role Is Not:

Qualification

A degree in Computer ScienceStrong PythonComfort with LinuxGrounding in the physical worldAn appetite for field workCamerasEmbedded Linux or edge computingRoboticsHardware-in-the-loop test rigsSimulation or game-engine work (GazeboC++ or TypeScriptDefense or aerospace experience

Required

We care more about strong engineering fundamentals and curiosity than about any specific technology or domain.
3+ years of professional software engineering, including owning a production system end to end: you designed it, shipped it, and fixed it when it broke.
A degree in Computer Science, Engineering, Physics or a related field, or equivalent experience.
Strong Python. You build services and tools other engineers depend on, not just scripts and notebooks.
Comfort with Linux, Docker, networking and CI. You can get a multi-service system running on hardware you've never touched before.
Real leverage from AI coding tools such as Cursor or Claude Code, with a considered view on what to delegate to them and what not to. We treat them as a multiplier on skills you already have: you should be able to write, explain and debug your own code without them.
Grounding in the physical world, whether from an engineering degree, hands-on work with hardware, sensors or robotics, or field time with systems that had to work outside the lab. You can read a schematic, understand a test rack, and talk to an optics or RF engineer.
An appetite for field work. You want to be on site when your code runs, and you can debug and ship a fix before the next flight.
A habit of measuring rather than guessing: you build the experiment, then trust the data.

Preferred

Cameras, EO/IR sensors, gimbals, radar, or video pipelines such as GStreamer or DeepStream.
Embedded Linux or edge computing, especially NVIDIA Jetson, and GPU or CUDA work.
Robotics, autonomous vehicles, drones or aerospace.
Hardware-in-the-loop test rigs, or test automation against real hardware.
Simulation or game-engine work (Gazebo, Isaac Sim, Unreal or similar).
Building AI agent harnesses or workflow graphs.
C++ or TypeScript.
Defense or aerospace experience, or a current U.S. security clearance.

Software Engineer, Sensing Systems at Chaos Industries · LumiJobs

Software Engineer, Sensing Systems

Responsibilities

Qualification

Required

Preferred

Benefits

Company

Chaos Industries

More roles at Chaos Industries