About

Harsheel Singh Soin is a highly skilled data scientist with a passion for machine learning and a strong background in Python, R, and SQL. With over 5 years of relevant experience, he has worked with a variety of libraries, including TensorFlow, Keras, Scikit-Learn, XGBoost, NLTK, and Gensim. Currently working at Audible, Inc. as a Data Scientist, Harsheel has spent the past 8 months working on projects such as predicting content quality metrics and studying the interaction between audiobook content and genre. In addition, he has worked with the Quantitative Investment Strategies team at Goldman Sachs, where he assessed pharmaceutical R&D similarity through FDA clinical trial metadata as part of his data science capstone project. As a student at Columbia University, Harsheel has taken courses such as Applied Machine Learning, Applied Deep Learning, Algorithms, and Exploratory Data Analysis. He has also worked on projects such as predicting first 6-month sales of audiobooks and studying the interaction between topics and genres. Harsheel enjoys the experimental process of model development, evaluation, optimization, and deployment, and has experience writing production-level code. He has worked as a Data Science Research Assistant at Amazon, where he built Gradient Boosting based regressors and classifiers to predict the first 6-month sales of audiobooks and their sales categories. He has also predicted content quality metrics using tuned Gradient Boosting Regressors based on writing-style and topic features, with the goal of identifying high-quality content for marketing, acquisition, and recommendation purposes. Harsheel holds a Master of Science in Data Science from Columbia University and a Bachelor of Technology (Honors) and Master of Technology in Biotechnology and Biochemical Engineering from IIT Kharagpur. He has also completed Intermediate Public Examination (Standard XII) from FIITJEE Junior College and All India Secondary School Examination (Standard X) from DAV Public School, Hyderabad. Overall, Harsheel is a highly skilled data scientist with a strong background in machine learning and a passion for solving complex problems.