Deep Learning Engineer (Computer Vision & Audio Analysis)

You will be joining a pioneering AI team where you will be responsible for designing and deploying cutting-edge deep learning solutions for computer vision and audio analysis. Your main tasks will include designing, developing, and optimizing deep learning models for image/video analysis (object detection, segmentation) and audio classification tasks. You will work with CNN architectures, Vision Transformers (ViT, Swin), and attention mechanisms (SE, CBAM, self/cross-attention) to address complex real-world challenges. In this role, you will process multi-modal data including video and audio. For video, you will apply spatiotemporal modeling (3D CNNs, temporal attention), while for audio, you will extract features (spectrograms, MFCCs) and build classification pipelines. You will also utilize pretrained models through transfer learning and multi-task learning frameworks, and optimize models for accuracy, speed, and robustness using PyTorch/TensorFlow. Collaboration with MLOps teams to deploy solutions into production is a key aspect of this role. Your required skills include advanced programming in Python (PyTorch/TensorFlow), expertise in computer vision concepts such as Vision Transformers, object detection techniques (YOLO, SSD, Faster R-CNN, DETR), and video analysis methods including temporal modeling. Additionally, you should have experience in audio processing, attention mechanisms, transfer learning, and training strategies. Experience in handling large-scale datasets and building data pipelines is also essential. Preferred qualifications for this role include exposure to multi-modal learning, familiarity with R for statistical analysis, and a background in publications or projects related to computer vision or machine learning conferences such as CVPR, NeurIPS, ICML. Please note that this position is for a client of Hubnex Labs, and selected candidates will work directly with the client's AI team while representing Hubnex.,