The Video Perception (VIPer) team develops perception systems for L2+ Advanced Driver Assistance Systems (ADAS). As an Applied Computer Vision Engineer, you will be involved in the full development lifecycle of the perception stack. You will work on real-world problems by analyzing large datasets, experimenting with Deep Learning models like Convolutional Neural Networks (CNNs) and Transformers, and contributing to production-ready software.
Responsibilities
鈥evelop, train, and validate deep learning models for perception tasks such as object detection, semantic segmentation, and lane detection.
鈥ontribute to the design and implementation of experiments, performing rigorous analysis of model performance and identifying failure modes.
鈥nalyze large-scale, unstructured video data to derive insights and assist in curating high-quality datasets for model training.
鈥ollaborate with the team to build and maintain robust MLOps pipelines for data processing, training, and deployment.
鈥mplement and optimize algorithms in Python, ensuring they meet the performance requirements for real-time embedded systems.
鈥ocument and present experimental results, architectural choices, and technical findings to the team and stakeholders.
Required Qualifications
鈥ducation: Bachelor鈥檚 degree in Computer Science, Electrical Engineering, or a related field.
鈥xperience: 3-8 years of professional experience in computer vision or machine learning application development.
鈥rogramming: Proficiency in Python and a strong understanding of object-oriented programming.
鈥eep Learning Frameworks: Strong hands-on experience with modern DL frameworks such as PyTorch or TensorFlow 2+.
鈥ore Concepts: Solid understanding of deep learning fundamentals, including CNNs, object detection, and segmentation. A keen interest in learning and applying Transformers for vision is essential.
鈥ools: Familiarity with essential software development tools like Git, Docker, and working in a Linux environment.