ZestIOT - Computer Vision Engineer - OpenCV/PyTorch

<p><p><b>Key Responsibilities :</b><br/><br/></p><p>- Lead the design and implementation of advanced computer vision algorithms.<br/><br/></p><p>- Architect, train, and optimize deep learning models for object detection, semantic segmentation, and real-time inference across diverse environments.<br/><br/></p><p>- Collaborate with cross-functional engineering, product, and research teams to integrate computer vision capabilities into scalable, production-grade systems.<br/><br/></p><p>- Drive data preprocessing strategies to ensure high-quality inputs for model training, with a focus on robustness and generalizability.<br/><br/></p><p>- Conduct rigorous experimentation, benchmarking, and performance analysis to continuously improve model accuracy and efficiency.<br/><br/></p><p>- Develop and optimize algorithms for low-latency, high-throughput inference pipelines using GPU acceleration.<br/><br/></p><p>- Stay ahead of emerging trends in computer vision, deep learning, and edge deployment translating research into practical solutions.<br/><br/></p><p>- Mentor junior engineers and contribute to technical reviews, architecture decisions, and roadmap planning.<br/><br/><b>Required Skills and Qualifications :</b><br/><br/></p><p>- Expert-level proficiency in Python with deep understanding of algorithms and data structures tailored to computer vision.<br/><br/></p><p>- Strong foundation in machine learning, neural networks, and image processing techniques.<br/><br/></p><p>- Extensive hands-on experience with PyTorch for building, training, and deploying deep learning </p><p>models.<br/><br/></p><p>- Proficiency in OpenCV for image manipulation, filtering, edge detection, and segmentation.<br/><br/></p><p>- Solid grasp of foundational deep learning architectures (e.g., CNNs, U-Nets, ResNets) and their practical deployment.<br/><br/></p><p>- Experience working with multi-spectral or hyper-spectral data, especially for detection and segmentation tasks.<br/><br/></p><p>- Familiarity with real-time inference frameworks such as ONNX Runtime and TensorRT, including optimization for GPU-based deployment.<br/><br/></p><p>- Experience integrating with Redis, RabbitMQ, and SQL databases for data streaming and messaging.<br/><br/></p><p>- Proficient in Docker and CUDA frameworks for containerized model deployment and GPU acceleration.<br/><br/></p><p>- Strong command of Linux environments, including scripting, debugging, and performance tuning.<br/><br/></p><p>- Proven ability to solve complex problems independently and lead technical initiatives within a team.<br/><br/></p><p>- Excellent communication skills for cross-functional collaboration and technical documentation.</p><br/></p> (ref:hirist.tech)