Set sensor simulation direction. Own the multi-quarter technical roadmap for radar, lidar, and thermal modeling — which fidelity gaps matter, which modalities we add, and how the work sequences. You'll be the company's authority here, and product strategy in this area will follow your read of where the industry is heading.

Build physically accurate sensor models. Radar as the primary focus — RF propagation, RCS, Doppler, multipath, antenna patterns, and raw pre-detection output — extending across lidar and thermal/LWIR. Real implementation work in modern C++ against a real-time rendering pipeline, deterministic and reproducible frame to frame.

Validate against real sensor data. Define the metrics and methodology that quantify how close our output is to measured reality, and drive model improvements from what that analysis tells you. We want to move from "this looks right" to "here is the correlation, here is the error distribution, here is what we fixed."

Model the hard conditions. Weather and environmental effects — rain, wetness, snow accumulation, low visibility — are where sensor fidelity gets interesting. Expect heuristics as well as first-principles physics, and the judgment to know which the problem calls for.

Set the standard for scientific rigor. Extend the validation discipline you bring on radar and lidar across the rest of the sensor suite, camera included. If we're going to call ourselves a sensor simulation company, the testing has to back it up.

Partner with perception and ML teams. Translate real-world perception failure modes into concrete, prioritized fidelity requirements — both internally and in technical conversations with customers evaluating our output.

Grow the team's depth. Provide technical direction to the engineers working in this area, and raise the bar on how the broader team reasons about sensor physics.

Use AI tooling actively. LLM-assisted coding and literature review meaningfully accelerate this kind of work. We expect fluency here — the bottleneck in this role is physical insight, not typing speed, and we want you spending your time on the former.

Experience. A track record of building simulation, sensor-modeling, or computational-physics software that was actually deployed and relied upon. We weight demonstrated physical insight above years served — a recent PhD with the right depth and a fifteen-year radar veteran are both credible here.

Physics-first foundation. Strong grounding in physics, applied physics, or electrical engineering — frequently an MSc or PhD, though we care about the depth of understanding rather than the credential. You can read a paper, work through the math, and turn it into working code.

Radar and RF depth. Radar signal processing, automotive or industrial radar domain experience, or a closely adjacent RF/electromagnetics background you're ready to apply to simulation. Candidates currently in radar algorithm roles who want to move into simulation are very much in scope.

Sensor physics breadth. Deep grounding in the physics of one or more sensing modalities — radar/RF, lidar, thermal/infrared, or electro-optical — and the underlying math: EM propagation, signal processing, radiometry, linear algebra.

Validation rigor. Demonstrated experience validating models against real measured sensor data and reasoning carefully about accuracy, error, and what a discrepancy is actually telling you.

Productive in C++. Our engineering team works primarily in modern C++. You don't need to be a language expert, but you must be genuinely effective at writing, debugging, and extending performance-sensitive code in a large multi-library codebase with a real build system and CI.

Technical leadership. Ability to set direction, not just execute it — to look at a domain, form a view on where it's going, argue for it, and then deliver against it.

Communication. Comfortable moving between a physics discussion with a researcher, an implementation discussion with a graphics engineer, and a roadmap discussion with product and executive stakeholders.

Raw radar simulation. Experience modeling pre-detection radar output — ADC-level, range-Doppler, or point-cloud-before-clustering — rather than object-level detections.

Thermal and LWIR. Thermal imaging or radiometric modeling background, including material emissivity and thermal scene modeling.

Commercial sensor simulation. Time spent building or working on commercial sensor simulation software, or inside a Tier-1 or OEM sensor team.

Real-time engines. Experience with Unreal Engine or a comparable real-time engine, or with bridging physics-heavy C++ against a rendering pipeline.

GPU and ray tracing. Physics engines, ray tracing, or GPU compute for large-scale sensor simulation.

Domain context. Autonomous vehicles, robotics, ADAS, or aerospace sensing.

Shipped meaningful improvements to our radar model that measurably close the gap against real sensor data, with the validation results to prove it.

Published a sensor simulation technical roadmap that engineering, product, and leadership align around.

Established a validation methodology and metric set that the team uses as the standard for judging fidelity.

Become a credible technical voice with customers evaluating our sensor output.

Equity

Employer-paid supplemental medical, mental health, dental, and vision benefits

Flexible paid vacation, sick time, winter shutdown, and 11+ holidays per year

Paid parental leave

Equipment budget to optimize your setup

Annual learning and development stipend