AI/ML Engineer (UAV Perception)

AI/ML Engineer (UAV Perception) Position: AI/ML Engineer (3–4 roles) Location: Aluva / Kochi, India (with field testing at UAV ranges) Department: UAV AI Division, Canary Digital.ai Role Overview As an AI/ML Engineer (UAV Perception) , you will design, train, and deploy computer vision and sensor fusion models for Canary Digital's UAV platforms. Your work will focus on object detection, anomaly detection, EO/IR fusion, and real-time inference on Jetson-class edge devices , enabling UAVs to autonomously detect threats, intrusions, and unusual occurrences. You will collaborate with UAV Systems Engineers and RF/Comms specialists to deliver an end-to-end AI-powered UAV solution for defense and security. Key Responsibilities Develop and deploy real-time computer vision models (object/person/vehicle detection, tracking, re-ID) on Jetson devices. Build change detection and anomaly detection pipelines for perimeter security and patrol missions. Integrate EO (RGB) + IR (thermal) sensor fusion models for day/night operations. Optimize models for low-latency inference using TensorRT, DeepStream, CUDA. Create synthetic datasets for rare threat scenarios (weapons, intrusions, tampering). Collect, annotate, and retrain models from field UAV test data . Collaborate with RF engineers for friend/foe beacon + CV hybrid detection . Implement MLOps workflows : versioned datasets, model deployment pipelines, performance monitoring. Support Weather Guard AI integration for weather-aware perception tuning. Required Skills & Experience 2–6 years of experience in AI/ML or Computer Vision engineering . Strong background in Deep Learning frameworks (PyTorch, TensorFlow). Proficiency in object detection/tracking architectures (YOLOv8, Detectron2, EfficientDet, SORT/DeepSORT). Familiarity with anomaly detection methods (autoencoders, clustering, unsupervised ML). Hands-on with NVIDIA Jetson platform (TensorRT, DeepStream, CUDA). Strong Python/C++ coding skills. Experience with dataset annotation, augmentation, and synthetic data generation . Preferred Background Prior projects in autonomous systems, robotics, or defense AI . Experience with AirSim/Gazebo simulation data pipelines . Knowledge of multi-sensor fusion (EO/IR, lidar, radar). Familiarity with real-time constraints in embedded AI systems. What We Offer Work on mission-critical UAV AI perception systems with national impact. Collaborate with UAV engineers, RF specialists, and defense partners. Access to Jetson dev kits, EO/IR sensors, UAV test flights . Competitive salary + performance-based incentives