About

Charan Sg is a Machine Learning Engineer at Amazon with a passion for building ML and deep learning models for various applications such as computer vision, natural language processing, and recommendation systems. He is also interested in applying regression and forecasting techniques to analyze raw data and predict the future. Charan has 7.87 years of relevant experience, including his current role at Amazon, where he works on AI, ML, and NLP Comprehend. Prior to Amazon, he worked as a Deep Learning Software Engineer at Intel, where he built and optimized deep learning models using Keras and Tensorflow for various business problems. He also contributed to adding submatrix-based convolution kernels to the Intel DirectML Metacommand driver stack, which increased performance by 10-30% per draw. Before joining Intel, Charan worked as a Research Assistant at the University of Southern California, where he forecasted product demand using ARIMA models to ensure adequate inventory at the USC Bookstore. He also used person detection models to count queue length at bookstore checkouts and generated heatmaps to recommend product arrangements for easy customer movement. Additionally, he employed logistic regression, neural network, and tree-based classifiers to predict whether a person would take a photograph with a pride statue at USC. Charan holds a Master's degree in Electrical and Computer Engineering from the University of Southern California and a Bachelor of Engineering in Electronics and Communication from M.S. Ramaiah Institute of Technology. He also completed a Pre-University course in Physics, Chemistry, Mathematics, and Biology from MES College of Science and an SSLC from Adichunchanagiri English High School. Charan's tech stack includes ML, deep learning, NLP, recommendation systems, computer vision, AI, Keras, Tensorflow, software engineering, and research science. He is a highly skilled Machine Learning Engineer with a strong track record of success in building and optimizing models for various applications.