Chandra Sekhar Pragada

• Developed and utilized data-centric training techniques to optimize deep learning models, achieving 2x faster convergence over baseline methods. • Built custom datasets and data pipelines, resulting in 60% faster model iteration. • Implemented advanced augmentation and sampling strategies, increasing model generalizability. • Achieved 45% convergence speed through data-centric training optimizations.

• Research, development, Troubleshooting, debugging and operation of computer vision pipelines. • Developed API for Automatic labelling and monitoring application with a user-friendly interface to track the system status. • Integrated the AI solution with hardware for real-time automated Quality inspection. • Deployed machine learning solutions onto edge devices. • Participated in Agile Scrum team, ensuring successful product delivery through daily stand-ups, sprint planning, retrospectives and reviews.

Deep Learning model for Optical Inspection of Oven-Printed Glass Flat panels. • This work focused on a defect analysis task that requires engineers to identify the causes of yield reduction from defect classification results, using CNN-based transfer learning method Enamel surface inspection of an Oven Cavity with neural networks • The approach of this project is to detect the defects in the oven cavity images using autoencoders which are very significant for anomaly detection.

• Developed an automatic surface defect detection for raw metals using CNN. • Developed, implemented and maintained quality control systems, processes, and procedures to ensure products and services meet company standards and customer requirements. • Developed and implemented a computer vision system for real-time defect detection on the production line for brake rotors, resulting in a 30% reduction in defects and a 10% increase in efficiency.

Subjects: Image Processing, Machine Learning, Statistics, Reliability of Mechatronic Systems.

Subjects: Manufacturing Automation, Non-Destructive Testing, Applied Numerical Methods, Statistics, Python.