We are seeking an AI Algorithm Engineer to develop and optimize core algorithms enabling autonomous flight operations for our UAV platforms. This role spans two critical algorithmic domains: computer vision and perception (detection, classification, tracking, SLAM, GNSS-denied navigation) and flight control state estimation (Kalman filtering, sensor fusion, pose estimation). You will design, implement, and validate algorithms that enable our drones to perceive their environment, navigate autonomously, and maintain stable flight in challenging conditions including GPS-denied and contested environments. This position requires both strong theoretical foundations and hands-on implementation skills to deliver production-ready, real-time algorithms for safety-critical autonomous systems.

Computer Vision & Perception Algorithms:
- Design and implement real-time detection, classification, tracking, and visual SLAM algorithms for autonomous navigation and GNSS-denied operations.
- Optimize deep learning models (YOLO, transformers, CNNs, or develop proprietary DL model) for edge deployment on embedded platforms (NVIDIA Jetson, Qualcomm RB5) under strict latency, power, and memory constraints.
- Develop vision-based pose estimation, optical flow, feature tracking, and change detection algorithms for mission-critical applications.

State Estimation & Flight Control Algorithms:
- Design and implement Kalman filters (EKF, UKF), complementary filters, and sensor fusion pipelines integrating IMU, GPS, vision, and RF positioning for robust state estimation.
- Develop motion prediction, trajectory estimation, and adaptive filtering techniques for dynamic flight conditions, sensor degradation, and environmental uncertainties.
- Create control-theoretic estimators for target tracking, collision avoidance, and swarm coordination.

Integration & Validation:
- Integrate algorithms with flight control units and companion computers; conduct benchmarking using simulation (Gazebo, MATLAB/Simulink) and HIL/HITL testbeds.
- Validate algorithms through field trials; analyze telemetry, measure KPIs (accuracy, latency, robustness), and iterate improvements.
- Document algorithm designs, performance characteristics, and integration interfaces; support certification efforts.

Required:
- Master's or PhD in Computer Science, Electronic Engineering, Robotics, Control Systems, or related field.
- 5+ years of work/academic experience developing algorithms for computer vision and/or autonomous systems (UAVs, robotics, autonomous vehicles) in real-world applications.
- Proven expertise in computer vision, including: object detection and tracking, feature extraction, SLAM (visual or sensor-fusion), visual odometry, and camera calibration.
- Solid knowledge of state estimation and control, including Kalman filter design (EKF, UKF), sensor fusion, and probabilistic motion models.
- Strong programming proficiency in C++ (Eigen, Ceres, GTSAM) and Python (NumPy, SciPy, pandas, scikit-learn);
- Advanced use of core computer vision libraries (OpenCV, scikit-image) and industry- standard deep learning frameworks (PyTorch, TensorFlow/Keras, ONNX).
- Hands-on experience with state-of-the-art detection and segmentation models (YOLOv8+/Detectron2/MMDetection/Mask R-CNN/U-Net, etc.) and modern feature extraction pipelines.
- Proficient in ROS2: package development, sensor/message integration, data pipelines.
- Demonstrated ability to deploy algorithms on embedded platforms (NVIDIA Jetson, Qualcomm RB5, ARM) under performance/reliability constraints.
- Strong mathematical foundation in linear algebra, probability theory, optimization, and numerical methods.
- Applied experience with simulation tools (MATLAB/Simulink, Gazebo, PX4 SITL) and direct hardware-algorithm integration/testing.
- Excellent collaborative skills; able to coordinate across software, hardware, and project teams; clear written and verbal technical communication.
- Robust knowledge of modern version control (Git/GitLab/GitHub) and dataset/model management (DVC or equivalents).
- Language proficiency: English - Upper intermediate or higher..

Preferred:
- Experience with swarm coordination algorithms, multi-agent systems, and distributed optimization.
- Knowledge of edge AI optimization and deployment: quantization/pruning/model compression for TensorRT, ONNX Runtime, OpenVINO, PyTorch/TensorFlow optimization toolkits.
- Experience with annotation platforms (Label Studio, Labelbox), experiment tracking (MLflow/Neptune.ai/TensorBoard), and visualization (Matplotlib, Seaborn, Plotly).
- Familiarity with flight control stacks (PX4/ArduPilot), MAVLink, and GNSS-denied navigation approaches (VINS, resilient localization).
- Track record of publishing in relevant conferences/journals (CVPR, ICRA, IROS, IEEE Transactions) or contributions to open-source repositories.