SPAICE is building the autonomy operating system that empowers satellites and drones to navigate and interact with the world – regardless of the environment. From GPS-denied zones on Earth to the unexplored frontiers of space, our Spatial AI delivers unprecedented levels of autonomy, resilience, and adaptability.
At SPAICE, you’ll work on real missions alongside leading aerospace and defense contractors, shaping the future of space and autonomous systems. If you're looking for a place where your work has a real, tangible impact – SPAICE is that place.
About the RoleSatellites that detect & avoid threats on their own. Drones that collaborate in GPS‑denied fields. Spacecraft that rendezvous with tumbling targets in orbit. All of these feats demand robust scene understanding from multiple sensors. That’s where you come in.
As a Computer Vision Engineer (Multimodal Sensing) you’ll implement and refine perception algorithms that fuse cameras, LiDAR, radar, event sensors, and beyond. Working shoulder‑to‑shoulder with a top‑tier team of CV scientists, you’ll translate cutting‑edge research into flight‑ready code for space and defense missions.
What you might work onImplement perception pipelines for situational awareness, collision detection & avoidance, formation flying, surveillance and terrain mapping across satellites and drones operating in GPS‑denied, dynamic environments.
Build the Perception stack of our Spatial AI architectures, fusing visual, inertial, and depth cues for robust, multi‑sensor scene understanding.
Integrate sensor fusion & neural representations to create dense onboard world models that run in real time on resource‑constrained hardware.
Deploy semantic scene understanding, visual place recognition, pose estimation, and monocular depth estimation on embedded or edge‑AI processors.
Collaborate with a top‑tier team of CV scientists and cross‑disciplinary engineers, delivering well‑tested, high‑performance code into flight processors, Hardware‑in‑the‑Loop (HIL) and Software‑in‑the‑Loop (SIL) setups, and real missions.
M.S. in Computer Vision/Robotics, or a related field.
Expertise in at least two of the following: multimodal perception & sensor fusion, neural representations, semantic scene understanding, SLAM / camera‑pose estimation, monocular depth estimation, visual place recognition.
Strong software engineering skills in C++ and Python, including performance‑critical CV/ML code on Linux or embedded platforms.
Familiarity with GPU or edge‑AI acceleration (CUDA, TensorRT, Vulkan, or similar).
Demonstrated ability to deliver production‑quality, well‑tested code in collaborative, fast‑moving environments.
Industry placements/working experience.
Familiarity with GPU or edge‑AI acceleration (CUDA, TensorRT, Vulkan, or similar).
Experience deploying perception pipelines on resource‑constrained hardware.
Publications in multimodal sensing/neural representations/SLAM for robotics or autonomous navigation in journals and conferences (e.g. CVPR, ICRA, ECCV/ICCV, NeurIPS).
Competitive salary commensurate with experience and impact.
Equity options – you will join us at the ground floor and share in the upside.
Well‑being perks – access to premium gyms, climbing centers, and wellness programs.
Team retreats & offsites – recent adventures include a half‑marathon in Formentera and a Los Angeles retreat during Oscars weekend.
Skills Required
- M.S. in Computer Vision, Robotics, or a related field
- Expertise in at least two areas including multimodal perception and sensor fusion, neural representations, semantic scene understanding, SLAM or camera-pose estimation, monocular depth estimation, and visual place recognition
- Strong software engineering skills in C++ and Python
- Experience developing performance-critical computer vision or machine learning code on Linux or embedded platforms
- Familiarity with GPU or edge-AI acceleration, such as CUDA, TensorRT, or Vulkan
- Ability to deliver production-quality, well-tested code in collaborative environments
- Industry placements or working experience
- Experience deploying perception pipelines on resource-constrained hardware
- Publications in multimodal sensing, neural representations, or SLAM for robotics or autonomous navigation
What We Do
SPAICE is an AI software company dedicated to Space & Defense, enhancing autonomous, real-time decision-making on edge for complex missions in harsh environments. They build spatial-AI autonomy software that turns aircraft, drones, and satellites into fully autonomous Spatial Agents.
