AI/ML Researcher: Computer Vision

Job Summary:

We are looking for an AI/ML Researcher to support the development of advanced artificial intelligence solutions for UAV (Unmanned Aerial Vehicle) and defence applications. The role involves researching, implementing, testing, and optimizing AI algorithms for image, video, navigation, and textual data representation in a rigorous, research-driven environment, under the guidance of senior engineers and domain experts.

Key Responsibilities:

• Apply classical computer vision techniques — including feature extraction, homography estimation, optical flow, camera calibration, and epipolar geometry — alongside modern deep learning approaches, selecting the most effective methodology for each problem domain.
• Design, develop, and maintain multi-object tracking systems integrated with object detection and re-identification (ReID) pipelines for robust real-time perception.
• Fine-tune and optimize deep learning models using advanced techniques such as knowledge distillation and Parameter-Efficient Fine-Tuning (PEFT), leveraging frameworks like Transformers while designing architectures based on CNNs and Vision Transformers (ViTs).
• Develop advanced 3D scene understanding capabilities, including Structure-from-Motion (SfM), visual odometry, and occupancy estimation using monocular, stereo, and depth-sensor data.
• Implement and maintain image-based 3D reconstruction pipelines for large-scale mapping, localization, and scene modelling from UAV imagery.
• Integrate perception outputs such as detection, segmentation, tracking, and depth estimation into downstream autonomous systems including path planning, obstacle avoidance, geo-referencing, and navigation modules.
• Train and optimize Reinforcement Learning (RL) models for autonomous guidance, navigation, and control of UAVs and aircraft operating in dynamic and mission-critical environments.
• Own the complete AI/ML model lifecycle, from dataset collection, curation, annotation strategy, and augmentation to training, evaluation, ablation studies, optimization, and deployment in production environments.
• Collaborate closely with firmware, robotics, autonomy, and software engineering teams to integrate computer vision and AI systems into real-world defence hardware platforms and flight-control stacks operating in mission-critical environments.

Required Technical Skills

• Strong hands-on experience with classical computer vision algorithms, including feature detection and description techniques (SIFT, ORB, AKAZE), optical flow methods (Farnebäck, Lucas-Kanade), and geometric transformations such as homography estimation, epipolar geometry, and essential/fundamental matrix computation.
• Proven expertise in building and deploying object detection, multi-object tracking, segmentation, and re-identification systems using modern architectures such as YOLO variants, DETR, SAM, Mask R-CNN, and related state-of-the-art frameworks in production-grade environments.
• Solid understanding of multi-view geometry and 3D vision concepts, including camera models, stereo rectification, disparity estimation, triangulation, pose estimation, and bundle adjustment.
• Experience with 3D data processing and reconstruction using libraries such as Open3D, PCL, or similar frameworks, including point cloud registration (ICP), filtering, segmentation, meshing, and spatial analysis.
• Familiarity with Structure-from-Motion (SfM), SLAM, and visual-inertial odometry frameworks (e.g., COLMAP, OpenMVG, ORB-SLAM, VINS-Mono) for UAV localisation, mapping, and autonomous navigation.
• Experience optimising AI/ML models for edge and embedded deployment using techniques such as quantisation, pruning, TensorRT acceleration, ONNX export, and runtime profiling.
• Strong understanding of sensor fusion methodologies, combining camera, IMU, GPS, LiDAR, and other onboard sensor data for robust perception, localisation, and navigation.
• Comfortable working with geo-referenced imagery, coordinate frame transformations, and geospatial mapping tools relevant to UAV and defence applications, including GeoTIFF, GDAL, and WGS84/UTM coordinate systems.
• Experience developing or training Reinforcement Learning (RL) models for autonomous guidance, navigation, control, and decision-making systems for UAVs, aircraft, or robotic platforms operating in dynamic environments.

Qualifications:

• Bachelor's degree in Computer Science, Robotics, or a related field
• At least one year of experience
• Understanding of computer vision and image processing concepts
• Familiarity with linear algebra fundamentals
• Participation in hackathons or robotics competitions
• Strong problem-solving attitude and willingness to learn

If you are passionate about computer vision and excited to work on real-world UAV and drone-based AI systems, this role offers a great opportunity to learn, build, and grow with an experienced engineering team.

Benefits:

• Lunch served at the office
• Full-time employees may be eligible for ESOPs according to the company's ESOP policy
• Health Insurance and PF
• Opportunity to work on innovative technology

About Us As the skies become increasingly crowded with UAVs, one startup is revolutionising defence with groundbreaking technology.

Introducing Zebu Intelligent Systems, a pioneering force in the defence sector dedicated to safeguarding our airspace. With the proliferation of unmanned aerial vehicles (UAVs), protecting the lower airspace has emerged as a critical global imperative. At Zebu, we're not just addressing this challenge – we're redefining it.

Our cutting-edge, patented counter-drone technology stands at the forefront of innovation, providing comprehensive protection against threats posed by nano, micro, small, medium, and large UAVs. As the global counter-drone market continues its exponential growth trajectory, poised to reach $11 billion by 2030, Zebu is positioned for unparalleled success.

Already, our solutions have garnered significant traction, with paid Proof of Concepts (POCs) secured from esteemed entities under the Ministry of Defense, Including DRDO, Indian Air Force, Indian Coast Guard and Tier 1 Aerospace and Defense companies.

Led by a visionary Ph.D. in Intelligent Systems and Robotics, our dynamic founding team is fueled by a shared passion for addressing the most pressing defence challenges of our time. If you're looking for a dynamic and rewarding career in AI and Robotics, we invite you to join our team at Zebu Intelligent Systems. We offer competitive compensation packages, opportunities for career growth, and a work environment that encourages innovation and creativity. Come be a part of our mission to change the world with intelligent solutions from Hyderabad, Telangana!

Culture We believe that our success is due in large part to our commitment to fostering a culture of innovation and collaboration. We believe in a hustling culture where we work hard to achieve our goals and make a meaningful impact. We value productivity and recognize that it takes dedication and effort to succeed. We are committed to creating a supportive work environment where everyone can thrive and achieve their full potential. We encourage creativity, innovation, and risk-taking, and we believe that everyone's ideas and contributions are important. We are dedicated to delivering results and are always looking for ways to improve and exceed expectations.

Equal Opportunity Our company is an equal opportunity employer and prohibits discrimination and harassment of any kind. We are committed to providing employment opportunities to all qualified individuals, regardless of race, colour, religion, sex, national origin, age, disability, gender identity or expression, sexual orientation, or any other legally protected status. We encourage applicants from diverse backgrounds to apply for our open positions and are committed to creating a welcoming and inclusive work environment for all employees.