AI/ML ADAS Perception Engineer
Magna International Inc.4 days ago
Auburn Hills, MI, USAMid Level
Responsibilities
- Develop, implement, optimize, and test conventional image-processing algorithms for automotive embedded serial-production projects.
- Design, tune, optimize, and validate deep learning models for ADAS computer vision features.
- Plan experiments, compare model architectures, analyze large datasets, and validate results through component and bench testing.
- Apply computer vision techniques including object detection, segmentation, tracking, image enhancement, lens distortion correction, homography, and projective transformations.
- Optimize algorithms and models for real-time execution and accuracy, latency, memory, and power targets on production ECUs.
- Collaborate with requirements, embedded, and testing teams to integrate models into automotive systems.
- Analyze defects and test results, perform root-cause analysis, and implement improvements to meet KPIs.
- Document development processes, maintain experiment logs, communicate findings, and independently deliver intermediate-to-advanced ADAS algorithm work.
- Follow automotive quality, safety, configuration-management, and software-development processes.
Requirements
- Master’s degree in computer engineering, software engineering, electrical engineering, computer science, or equivalent.
- At least 3 years of experience developing computer vision and image-processing algorithms using traditional methods and deep learning.
- Proven expertise developing and implementing deep neural network models.
- Excellent programming skills in C or C, with familiarity with Python and deep learning frameworks such as TensorFlow, PyTorch, and Keras advantageous.
- Knowledge of machine learning concepts and neural network architectures including CNNs, RNNs, and transformers.
- Experience with image segmentation, object detection, image-data preparation and enhancement, normalization, augmentation, filtering, noise reduction, contrast adjustment, and image restoration.
- Experience optimizing models for performance using techniques such as quantization and distributed training.
- Strong foundations in mathematics, signal/image/video processing, computer vision, and embedded AI development.
- Experience developing portable, reusable, modular software for automotive systems and familiarity with automotive product-development practices.
- Strong troubleshooting, debugging, structured problem-solving, communication, and independent-working abilities.
- Experience with serial development of automotive electronics is preferred.
- Familiarity with ASPICE or CMMI, configuration management, and project monitoring and control techniques is required.
- Comprehensive spoken and written English is required.
Benefits
- Engaging and dynamic environment focused on industry-leading automotive technologies.
- Employee support and development resources with varied career-development prospects.
- Regular/permanent employment with Magna Electronics.
- Global, multidisciplinary work environment.
- Equal opportunity employment and accommodations for individuals with disabilities.
Categories