We're Hiring: Computer Vision Specialist
Full time | Core team
LAT Aerospace | Perception & Autonomy
At LAT Aerospace, we’re building India’s first clean-sheet hybrid-electric STOL aircraft and a next-generation autonomy stack powering aircraft for complex missions. Our team is kicking off a new chapter — developing the full flight-control, obstacle-avoidance, GNSS-denied navigation, and swarm-coordination layers that will define LAT’s autonomy architecture.
You will design and implement real-time CV algorithms for extracting features, tracking structure, and generating robust visual cues for localization, mapping, and inter-drone relative positioning. This role sits at the heart of our autonomy architecture.
What you’ll do
Develop real-time computer vision algorithms using OpenCV and custom geometric vision pipelines for feature detection, feature tracking, keypoint extraction, and visual correspondence.
Build robust visual processing modules for edge devices — capable of handling motion blur, lighting variation, small textures, low-altitude flight perspectives, and high-dynamic maneuvers.
Extract structural cues (edges, corners, lines, descriptors) that drive VIO, SLAM, and GNSS-denied navigation pipelines.
Implement multi-view geometric vision including epipolar geometry, homography estimation, essential matrix computation, and depth inference from motion.
Optimize CV algorithms for embedded Linux systems and edge compute: SIMD, ARM optimizations, memory/latency efficiency.
Develop visual debugging tools for inspection of keypoints, flow fields, descriptors, and algorithmic performance in real flight datasets.
What we're looking for
Strong experience in classical computer vision, particularly feature detection, extraction, tracking, and geometric vision.
Expertise with OpenCV, custom CV pipelines, and real-time image processing.
Strong knowledge of epipolar geometry, camera models, triangulation, motion estimation, or similar domains.
Strong Python and/or C++ experience for vision algorithms.
Experience optimizing algorithms on embedded hardware, GPUs, or constrained compute environments.
Ability to evaluate algorithm robustness on real datasets and noisy field conditions.
Why LAT?
Core-team role building the autonomy brain of India’s most ambitious aerospace startup.
Ownership, speed, and the opportunity to define a new class of systems from scratch.