About this opportunity
Urban Sky lists this Senior Embedded Imaging Software Engineer opportunity in northern, Kentucky. Review the employer’s description below for duties, qualifications and application requirements.
Job description
Senior Embedded Imaging Software Engineer
About Urban Sky
Urban Sky is a venture-backed aerospace startup on a mission to pioneer humanity’s routine, easy access to the stratosphere and the value locked within it. We design, build, operate, and sell stratospheric balloon systems. The data we collect supports a wide range of applications, from real-time wildfire monitoring to urban mapping. Our technology is also employed by government customers including the Department of War (DoW), NASA, and others. Urban Sky sends flight vehicles to the stratosphere on a weekly basis.
Our team consists of engineers, makers, developers, and doers who believe in the power of human potential when passion meets utility within a small, supportive team. At Urban Sky we strive to create a team culture grounded in candor, inclusion, respect and empowerment.
We aim to build the next chapter of stratospheric history in Denver and Texas and are seeking self-starting, open-minded, and hard-working team members.
Urban Sky is seeking a SeniorEmbeddedImaging Software Engineer to own the firmware, imaging pipeline, and data-movement stack behind our electro-optical (EO) and infrared (IR) payloads – from the sensor at the focal plane to the command-and-control link on the ground.
You’ll write the firmware that drives our cameras and edge compute, own the CUDA image processing pipeline that turns raw pixels into clean, calibrated, geo-tagged imagery, and architect the high-rate data paths and C2 infrastructure that connect the payload to the customer. Your goal: deliver clean, honest imagery. Much of this work proves out in the field – at events, exercises, and customer engagements.
Key Responsibilities:
Own embedded firmware for our EO and IR payloads: sensor bring-up, driver development, and edge compute on NVIDIA Jetson platforms
Integrate cameras and sensors over PCIe, MIPI/CSI, and USB, and develop board-support packages for custom carrier boards, system-on-modules, and companion microcontrollers
Own the L0 image processing pipeline running in CUDA at the edge: debayering, white balance, gamma, and camera/lens correction – applied to raw frames before any codec
Own IR-specific processing: non-uniformity correction (NUC), bad-pixel replacement, and radiometric calibration
Develop multi-frame super-resolution and related computational imaging capabilities
Implement imagery metadata (KLV/MISB) so every frame carries accurate pointing, position, and time
Architect high-throughput data paths across the vehicle – PCIe as a first-class specialty – and the routing infrastructure that carries imagery, telemetry, and C2 across satellite, tactical-radio, and IP links
Optimize for latency, throughput, and determinism, and balance pipeline compute against onboard ML workloads sharing the same CUDA resources
Integrate payload firmware with the gimbal and motion-control stack, consuming pointing and autonomy commands from the broader payload software architecture
Develop and maintain optical calibrations for our EO/IR payloads – lens distortion modeling, vignetting/flat-field correction, and MTF/sharpness characterization – so imagery quality is measured and repeatable across cameras and flights
Implement motion and temporal compensation (frame-to-frame registration, stabilization, and temporal filtering) to maximize effective image quality from a moving platform
Qualifications:
Experience building end-to-end camera/imaging systems, from sensor to downstream consumer
Working experience with image processing pipelines (debayering, calibration, correction) and CUDA or GPU-accelerated processing
Experience with optical calibration and characterization – distortion modeling, vignetting correction, MTF/sharpness measurement – and building the correction steps that apply them in a processing pipeline
Familiarity with motion/temporal compensation techniques (registration, stabilization, temporal filtering) for imaging from moving platforms
5+ years developing embedded firmware in C/C++ for embedded Linux and/or bare-metal/RTOS systems
Hands-on experience shipping on NVIDIA Jetson or comparable embedded Linux/SoC platforms
Deep experience moving high-rate data over PCIe (driver, DMA, endpoint/root-complex level)
Solid grounding in networking, data routing, and command-and-control infrastructure
Strong hardware debugging skills and comfort working from schematics
Nice-to-Haves / Preferred Skills:
IR/thermal imaging experience, including NUC and radiometric calibration
GStreamer, V4L2, or other imaging/video pipeline frameworks
Computational imaging techniques (super-resolution, stacking, HDR)
Defense imagery and metadata standards (MISB/KLV, STANAG); defense or mission-critical development experience
FPGA workflows or high-speed serial interfaces (SerDes, Aurora, MIPI)
Experience with optics and optical hardware
Experience with multi-sensor and multispectral fusion – co-registering and combining imagery across sensors and bands into fused data products, in partnership with downstream data teams
Requirements:
Must be US Citizen or Permanent Resident
Must be able to obtain and maintain a US security clearance (an active clearance is a plus)
Must be located near Denver, CO or willing to relocate; role is on-site at our Denver headquarters (4800 Race St)
Travel expected to support field events, exercises, and customer engagements
Stock Options
Medical, Vision and Dental
Flexible PTO
Urban Sky is an Equal Opportunity employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected Veteran Status, or any other characteristic protected by applicable federal, state, or local law.
The pay range for this role is:
140,000 - 180,000 USD per year (Denver CO HQ)
#J-18808-Ljbffr
Who can apply
Review the original listing for work authorization, qualifications and employer requirements.