Varun Sivasubramanian
About Varun Sivasubramanian
Varun Sivasubramanian is a Senior Data Scientist with extensive experience in demand forecasting and machine learning. He has worked at various organizations, including Impact Analytics and Reliance Industries Limited, and holds degrees in Chemical Engineering and Artificial Intelligence.
Work at Impact Analytics
Currently, Varun Sivasubramanian serves as a Senior Data Scientist at Impact Analytics, a position he has held since 2020. In this role, he focuses on developing advanced data-driven solutions to enhance business decision-making. His work includes generating optimal promotion and pricing strategies for clients, utilizing forecasting models to predict demand accurately. He has also contributed to the incorporation of a store split module in an AIML-based pipeline, which aids in generating store-level forecasts.
Education and Expertise
Varun Sivasubramanian holds a Post Graduate Program in Artificial Intelligence and Machine Learning from The University of Texas at Austin - Red McCombs School of Business, completed from 2019 to 2021. He also pursued a similar program at Great Learning during the same period. Prior to this, he earned a Bachelor’s Degree in Chemical Engineering from the National Institute of Technology Karnataka, graduating in 2017. His educational background equips him with a solid foundation in data science and engineering principles.
Background
Varun Sivasubramanian began his career as a Summer Intern at the Council of Scientific and Industrial Research in 2015. He later interned at JSW Steel in 2016 for one month. Following his internships, he worked as a Process Engineer at Reliance Industries Limited from 2017 to 2019. His diverse experience in engineering and data science has contributed to his current expertise in demand forecasting and data analysis.
Achievements
During his tenure as a Data Scientist at Impact Analytics from 2019 to 2020, Varun Sivasubramanian built an ensemble forecasting model for a US-based retail-storage-solutions chain, addressing issues related to lost sales and overstocking. He also developed an advanced algorithm for a demand forecasting pipeline that selects the best model based on multiple error metrics. Additionally, he analyzed changes in footfall before and after lockdowns to create a mobility factor for updating demand predictions in the post-COVID period.