About

Prateek Chandra Jha is an Applied Scientist at Amazon, where he has been instrumental in shaping machine learning (ML) models as part of the company's Applied Science charter. With a background in aerospace engineering, Prateek has also worked as an Applied Machine Learning Scientist at CognitiveScale and a Deep Learning Engineer at Hopstack. At Hopstack, Prateek collaborated with the engineering team to implement state-of-the-art neural architectures for inertial navigation inside supply chain warehouses, involving estimation of positions and orientations of a moving subject from a sequence of IMU sensor measurements. He implemented key neural network models like ResNets, Temporal Convolutional Networks (TCN), and Bilinear LSTMs, making continuous improvements for challenging motion edge cases. Prateek holds a Bachelor's degree in Aerospace Engineering from IIT Bombay, a Master's degree from Chennai Mathematical Institute, and a Master of Science in Financial Engineering from World Quant University. He also has certifications in CBSE Senior Secondary Certificate Exam and ICSE Certificate Exam - 10th Boards. Prateek's technical skills include research scientist, machine learning, deep learning, and neural networks. With 2.12 years of relevant experience, he continues to deliver work on the mathematical machinery behind ML models, subsequent optimization, deployment, and working of ML models in production.