Computer Vision Engineer

We are seeking a proactive Computer Vision Engineer who excels in dynamic environments and has a passion for developing practical AI systems. If you have a keen interest in working with video, visual data, cutting-edge ML models, and resolving impactful challenges, we are eager to connect with you. This position merges deep learning, computer vision, and edge AI, focusing on constructing scalable models and intelligent systems to drive our advanced sports technology platform. Your responsibilities will include designing, training, and refining deep learning models for real-time object detection, tracking, and video comprehension. You will be responsible for implementing and deploying AI models utilizing frameworks such as PyTorch, TensorFlow/Keras, and Transformers. Working with video and image datasets using tools like OpenCV, YOLO, NumPy, Pandas, and visualization tools like Matplotlib will be a key aspect of your role. Collaborating closely with data engineers and edge teams to deploy models on real-time streaming pipelines will also be part of your duties. Additionally, you will need to optimize inference performance for edge devices such as Jetson and T4, and manage video ingestion workflows. You will also be expected to rapidly prototype new concepts, perform A/B tests, and validate enhancements in real-world scenarios. Clear documentation of processes, effective communication of findings, and contributing to the expansion of our AI knowledge base are essential aspects of this role. To be successful in this position, you should possess a strong command of Python and have familiarity with C/C++. Experience with deep learning frameworks such as PyTorch, TensorFlow, and Keras is required. A solid understanding of YOLO, Transformers, or OpenCV for real-time visual AI is essential. Proficiency in data preprocessing, feature engineering, and model evaluation using NumPy, Pandas, etc., is also necessary. A good grasp of computer vision, convolutional neural networks (CNNs), and object detection techniques is expected. Exposure to video streaming workflows like GStreamer, FFmpeg, and RTSP will be advantageous. The ability to write clean, modular, and efficient code is crucial for this role. Experience in deploying models in production, particularly on GPU/edge devices, is highly valued. An interest in reinforcement learning, sports analytics, or real-time systems will be considered a plus. An undergraduate degree in Computer Science, Artificial Intelligence, or a related field is required, while a Master's or PhD is preferred. A strong academic background will be beneficial for this position.,