Prashanth Dannamaneni
About Prashanth Dannamaneni
Prashanth Dannamaneni is a Data Scientist at Pinsight Media in Kansas City, Missouri, with eight years of experience in the field. He holds a Master's degree in Statistics from the University of Michigan and a Bachelor's degree in Civil Engineering from the National Institute of Technology Warangal.
Work at Pinsight Media
Prashanth Dannamaneni has been employed at Pinsight Media since 2016 as a Data Scientist. In this role, he has contributed to various projects aimed at optimizing advertising campaigns. He previously served as a Data Science Intern at the same company in 2015 for seven months. His work has included developing a rule-based model that resulted in a 60% improvement in click-through rates while maintaining a lower cost-per-click compared to untargeted groups in tested campaigns.
Education and Expertise
Prashanth Dannamaneni holds a Master's degree in Statistics from the University of Michigan, where he studied from 2014 to 2015. He completed a Bachelor's degree in Civil Engineering at the National Institute of Technology Warangal from 2010 to 2014. His educational background provides a strong foundation in analytical skills and quantitative methods, which he applies in his data science work.
Background
Before joining Pinsight Media, Prashanth worked as a Research Assistant at the Institute of Social Research at the University of Michigan for three months in 2014. This role contributed to his experience in data analysis and research methodologies, further enhancing his capabilities in data science.
Technical Skills and Projects
Prashanth Dannamaneni has demonstrated proficiency in managing large datasets and implementing data processing techniques. He scheduled AWS data pipelines to handle over 3 billion raw bid requests daily using Spark on EMR. He has also utilized tower information and device interaction variables in logistic regression models and implemented feature hashing to expand feature space for model training. Additionally, he created a confidence radius model to predict accuracy bins for location events generated by 60 million Sprint subscribers.