Ans S.

AI Engineer specializing in Computer Vision, focused on building scalable systems that transform visual data into real-world applications. My work centers on developing and optimizing deep learning models for object detection, segmentation, tracking, and keypoint detection, particularly for edge environments. I have hands-on experience building models using TensorFlow and PyTorch, working with modern architectures such as YOLO, DETR, ResNet, EfficientNet, and CenterNet. My work also includes zero-shot object detection and segmentation for IP camera streams. I frequently optimize and export models using ONNX and TensorFlow Lite (TFLite) to enable efficient deployment on edge and mobile devices, ensuring low latency and reliable performance in production environments. My primary interest lies in designing scalable Edge AI solutions that bring computer vision capabilities directly to mobile and embedded platforms, enabling intelligent real-time applications.