Computer Vision Engineer

Key Responsibilities Design and implement novel high-precision computer vision and machine learning models for real-time road and cabin scene analysis such as object detection/segmentation/tracking, head/body pose estimation, event reconstruction and scene understanding, biometric recognition, camera calibration, sensor fusion, etc.Optimize models for accuracy and performance, focusing on reducing inference time and memory consumption with quantization, pruning, and knowledge distillation to reduce model size and maintain performance.Research and implement advanced techniques in deep learning, including but not limited to CNNs, RNNs, Transformers, Generative Models, and attention mechanisms.Work on embedded systems with hardware accelerators.Collaborate with hardware teams to align software solutions with hardware capabilities and constraints.Set up monitoring tools for edge-deployed models, ensuring their reliability in the field.Automate model update pipelines for seamless continuous integration and deployment in edge environments.Continuously fine-tune and improve the model performance based on live data.Collaborate with cross-functional teams including camera software engineers and platform developers to align on project goals and execution. Education: Bachelor's or Master's degree in Computer Science, Machine Learning, Artificial Intelligence, Mathematics, or a related field (Ph D is a plus). Experience: 3+ years of experience in computer vision, machine learning, or related fields, with a focus on high-precision models and edge deployment.Proven track record of deploying computer vision models on product edge devices or in resource-constrained environments.Familiarity with edge deployment tools and libraries such as Py Torch Mobile, Tensor Flow Lite, ONNX Runtime, Open VINO, etc. Technical Skills: Proficiency in Python, C++, and relevant libraries/frameworks (e.g., Py Torch, Tensor Flow, Keras, Open CV).Experience with hardware accelerators for edge deployment (e.g., Ambarella CVFlow, Qualcomm SNPE, NVIDIA Jetson).Deep understanding and hands-on experience in the neural network model compression and acceleration techniques such as quantization, pruning, knowledge distillation, and architecture optimization.Experience with real-time inference and low-latency processing for computer vision applications. Other Skills: Strong problem-solving skills and an ability to work independently or as part of a team.Ability to translate business requirements into technical solutions.Excellent communication skills with an emphasis on explaining technical concepts to non-technical stakeholders. #J-18808-Ljbffr