This role is contributing to the Robotics & Discrete Automation in Bangalore India.

• Design, train, and optimize deep learning models for core perception tasks—object detection, semantic and instance segmentation, and 3D scene understanding—using multi-modal sensor data (e.g., RGB-D).
• Devise labeling strategies, and tooling for automated annotation and QA. Contribute to dataset curation, balancing, and advanced augmentation techniques to simulate real-world illumination, occlusion, and high-mix variation scenarios.
• Develop continuous monitoring frameworks to assess model performance and reliability in real-world and simulated environments. Implement uncertainty estimation and failure detection methods to support fallback and recovery strategies in dynamic, unstructured conditions.
• Conduct proof-of-concept experiments to evaluate new algorithms, tools, and methodologies in computer vision, simulation, and perception. Translate research insights into practical prototypes that can inform roadmap and architecture decisions.

Qualifications for the role

• MS or PhD in Computer Vision, Robotics, Machine Learning, or a related technical field or equivalent hands-on experience building and deploying perception systems in real-world robotics applications.
• 6 to 8+years of hands-on experience building deep learning models for perception tasks such as object detection, segmentation, and 3D understanding.
• Strong proficiency with Python, C++, and deep learning frameworks like PyTorch or TensorFlow; experience with ROS2 is a plus.
• Deep understanding of multi-modal sensor data (e.g., RGB-D, LiDAR) and experience working with synthetic data, real-world datasets, and augmentation strategies for robustness.
• Experience optimizing models for real-time inference, including latency reduction, throughput scaling, and deployment in closed-loop robotic systems. Skilled in evaluation methodologies, diagnostics, and performance monitoring across both simulation and real-world environments.
• Solid understanding of robustness and reliability techniques, including uncertainty estimation, perception failure detection, and recovery strategies for dynamic, unstructured environments.
• Strong problem-solving skills, ability to work independently, and experience collaborating in cross-functional robotics teams.