Senior Radar System Engineer

At The ReWork Group, we partner with high-growth startups and forward-thinking companies to build the future.

Our client is building a platform where AI doesn't just analyze data; it reasons across it, forecasts what's coming, and acts!

As the Senior Radar System Engineer , you know Commercial SAR constellations are throwing off more data than anyone has figured out how to use. Our client is building the algorithms that find what's moving, where, and why it matters. If you want your work flying in production instead of sitting in a paper, keep reading.

Most radar talent spends its career one step removed from impact: a detector that works in a notebook, a result that clears review, a model nobody ever puts in front of an operator. We're the opposite.

The gap between "interesting" and "operational" is exactly where this role lives, and closing it is the job.

What you'll do:

• Build the detection stack. Design and ship GMTI and change-detection algorithms that turn multi-look SAR returns into moving-target tracks, in production, on a live constellation.
• Own the pipeline end to end. From raw phase history and complex imagery through detection, geolocation, and track association. No throwing prototypes over the wall to a separate "productionization" team; you carry it all the way.
• Fuse across the constellation. Exploit multi-pass, multi-sensor, and multi-modal data to knock down false alarms and hold tracks through gaps and revisit windows.
• Close the loop on real signals. Validate against ground truth, tune for clutter and terrain, and measure yourself on operational accuracy rather than benchmark scores.
• Ship fast, then harden. Get a working detector in front of users in weeks, then push latency, throughput, and reliability until an operator can trust it without thinking about it.

Who you are:

• You've lived in the signal. SAR, radar signal processing, or a close cousin (RF, sonar, seismic, EO/IR). You know complex-valued data is not just "images."
• You ship production code , not just research artifacts. Python plus C++, CUDA, or the equivalent when performance demands it.
• You've owned something real. You've had code in production and felt the consequences when it broke, or you're hungry to.
• You optimize for the field, not the reviewer. Whether the thing works where it's deployed matters more to you than whether it would survive peer review.
• Bonus points for GMTI or STAP, Kalman or particle filtering, deep learning on radar data, or eligibility for a security clearance.