Computer Vision Intern (Deep Learning)

idle Robotics, a Bengaluru-based startup, is on a mission to be the intelligence layer empowering all autonomous systems. You will play a crucial role in developing and training deep learning architectures like YOLO, Faster RCNN, and UNet for perception tasks. Additionally, you will assist in adapting foundation models such as CLIP, DINO, and SAM for robotic use cases. Your responsibilities will also include maintaining experiment logs, conducting failure analysis on model performance metrics, and applying TorchVision and Albumentations within Python workflows for model augmentation. Qualifications: - Technical proficiency in PyTorch or TensorFlow frameworks - Rigorous understanding of probability, calculus, and linear algebra - Exposure to vision-language or self-supervised models is advantageous As part of idle Robotics, you will collaborate directly with IIT/IISc founders in a high-density engineering environment. You will have the opportunity to solve complex "zero-to-one" problems in Physical AI and autonomous systems. Furthermore, you will contribute to developing dual-use technology for national defence (GPS-denied navigation) and industrial automation. Enjoy competitive compensation, paid time off, and immerse yourself in a growth-focused culture at idle Robotics. idle Robotics, a Bengaluru-based startup, is on a mission to be the intelligence layer empowering all autonomous systems. You will play a crucial role in developing and training deep learning architectures like YOLO, Faster RCNN, and UNet for perception tasks. Additionally, you will assist in adapting foundation models such as CLIP, DINO, and SAM for robotic use cases. Your responsibilities will also include maintaining experiment logs, conducting failure analysis on model performance metrics, and applying TorchVision and Albumentations within Python workflows for model augmentation. Qualifications: - Technical proficiency in PyTorch or TensorFlow frameworks - Rigorous understanding of probability, calculus, and linear algebra - Exposure to vision-language or self-supervised models is advantageous As part of idle Robotics, you will collaborate directly with IIT/IISc founders in a high-density engineering environment. You will have the opportunity to solve complex "zero-to-one" problems in Physical AI and autonomous systems. Furthermore, you will contribute to developing dual-use technology for national defence (GPS-denied navigation) and industrial automation. Enjoy competitive compensation, paid time off, and immerse yourself in a growth-focused culture at idle Robotics.