Senior AI Engineer (Efficient ML)

Your Mission

This position is responsible for designing, developing, fine-tuning, and optimizing deep learning models for robotics and quality inspection applications. You will contribute to the development of data-efficient machine learning (ML) algorithms for object detection, recognition, segmentation, and pose estimation across multiple modalities, including images, 3D, and multispectral sensors.

What To Expect

• Design, develop, and optimize deep learning models (e.g., convolutional neural networks, transformers) to process and analyze sensor data from various modalities (e.g., vision, depth, polarized light, patterned light) for anomaly detection and object segmentation.
• Research and develop data-efficient machine learning methods, such as zero-shot learning, few-shot learning, and transfer learning.
• Evaluate and calibrate various sensors to ensure the collection of accurate data for training.
• Pre-process and prepare image and sensor data for effective model training, including data cleaning and augmentation.
• Validate the performance of implemented algorithms on real-world car part datasets.
• Collaborate with robotics engineers to integrate AI solutions into robotic systems and ensure seamless operation.
• Document technical reports, program codes, and set-up manuals.

What You'll Bring

• A Master’s or Ph.D. degree in Electrical, Mechanical, or Computer Engineering, or a relevant discipline, with more than 5 years of industry experience.
• Familiarity with object detection, recognition, and segmentation frameworks (e.g., YOLO, Faster R-CNN, Mask2Former).
• A proven track record in efficient ML techniques, including zero-shot learning, few-shot learning, and transfer learning.
• Knowledge of advanced data augmentation and synthesis for 2D/3D sensors.
• Understanding of data quality requirements for different AI training tasks and data acquisition protocols.
• Familiarity with data preprocessing and augmentation methods for images, video, 3D point clouds, and other sensor data.
• Working knowledge of sensor methodologies, including data analysis, noise reduction, feature engineering, sensor calibration, and optimization strategies.
• Ability to evaluate and shortlist candidate solutions or sensors based on technical feasibility, performance, and cost.
• Knowledge of pose estimation, grasp planning, robot motion planning algorithms, and robotic simulation platforms.
• Proven experience in developing and deploying deep learning models in real-world applications.
• Proficiency in programming industrial robots using manufacturer-specific languages or general-purpose languages like C, C#, C++, Python, etc.
• Any patents, publications, specialist certifications, or awards in multimodal processing are an added advantage.