Data Scientist / AI-ML Engineer

About this role

About the role: - Join Advance Sensing to develop and deploy machine learning and artificial intelligence systems for industrial applications, with a focus on computer vision, defect detection, and automated quality inspection in manufacturing environments. - You will work on building, training, and optimizing models to analyze images and sensor data from production lines, identifying anomalies and quality issues at scale. - Your work spans the full lifecycle from data preparation through model deployment, including experimenting with architectures, tuning hyperparameters, and ensuring models perform reliably in production. What you'll work on: - Computer vision and image processing tasks using OpenCV and deep learning frameworks to solve real-world manufacturing challenges. - Classification, regression, clustering, and object detection workflows tailored to industrial inspection and quality assurance. - Convolutional neural networks and image-analysis architectures for defect detection, including semantic and instance segmentation for pixel-level identification of manufacturing faults. - Anomaly detection systems that flag previously unseen or rare defects that traditional methods might miss. - Time-series analysis and predictive analytics on industrial sensor data to forecast equipment issues or process drift. - Model optimization, quantization, and deployment strategies for edge AI systems that run on resource-constrained hardware in manufacturing plants. - Transfer learning approaches, leveraging pretrained models and fine-tuning them for your specific inspection tasks. - Data pipeline work, including dataset annotation, class balancing, image augmentation, and API integrations to feed models with production data. - Experiment tracking, model versioning, and deployment monitoring to maintain model quality over time. What we're looking for: - Strong foundation in AI, Machine Learning, and Deep Learning fundamentals, with the ability to explain concepts like supervised versus unsupervised learning, overfitting, regularization, and performance metrics. - Proficiency in Python for data science and engineering, with hands-on experience manipulating and analyzing data at scale. - Solid working knowledge of NumPy, Pandas, Matplotlib, and Scikit-learn for data wrangling and visualization. - Basic competence in PyTorch, TensorFlow, or Keras for building and training neural networks. - Familiarity with OpenCV and the core principles of image processing and computer vision. - Strong analytical and mathematical reasoning ability, with excellent debugging and systematic problem-solving skills. - Demonstrated interest in or passion for industrial AI, smart manufacturing, machine vision, and quality inspection automation. Preferred experience: - Production experience with computer vision and industrial image-processing applications. - Track record building CNNs for image classification, defect detection, or related tasks. - Hands-on work with object detection architectures and their deployment workflows. - Experience implementing semantic or instance segmentation to isolate defective regions at the pixel level. - Familiarity with anomaly detection techniques in industrial or manufacturing contexts. - Portfolio or examples of transfer learning and model fine-tuning on domain-specific datasets. - Experience with dataset labeling, addressing class imbalance, and applying image augmentation strategies. - Knowledge of model compression and edge deployment, including quantization and optimization for inference. - Background in time-series forecasting or predictive analytics for sensor or IoT data. - Working knowledge of SQL, relational databases, RESTful APIs, and data pipeline design. - Understanding of MLOps practices, including experiment logging, model registry, continuous integration for ML, and production monitoring.