Jul 2023 – Mar 2025 · 1 yr 8 mos · On-site

Apr 2021 – Aug 2023 · 2 yrs 4 mos · On-site

• Created a number of reports and a dashboard for Migration of 3.3 Crore users on Payzapp 1.0. Incorporated error codes which helped to reduce the migration failures by over 25 Percent.
• Supported Operations Team by automating a number of reports to facilitate transaction reconciliation. Optimised and converted report queries to incre- mental models which reduce the time to load under 1 min.
• Helped the product team identify gaps in offus transaction authentication and onboarding flows by creating funnels using relevant event data from google bigquery and kibana logs.
• Used Natural Language processing techniques(N-Gram Analysis) to find chal- lenges faced by users on neobank apps using review data and comments on social media.
• Used transactional & Demographic data from PZ 1.0 to validate hypothesis around user behavior for feature developments on payzapp. Used Clustering techniques and tool like metabase.
• Built a model to predict customer churn with 87 percent accuracy using Random forest Classifier in Directi – Data Science Hackathon. Applied Cleaning, feature engineering selection and model selection in the process.

Jun 2019 – Mar 2021 · 1 yr 9 mos · Gurgaon, India · On-site

• Helped in the development of package eta engine by incorporating average route, model path and processing time models.
• Migrated the package eta engine to use redis instead of querying athena and enhanced the accuracy by using the output of trucks eta engine.
• Helped the Operations Team better plan the peak season loads by using the output to predict the loads at subsequent centers.
• Extracted and Cleaned the training data based of a set of rules on GPS tracking data. Also, helped in the eta prediction logic.
• Optimised trucks-eta and reduced training run-time from 4 to 1.5 hours by reducing redundancy in data retrieval and cleaning processes. Built a predictive model for trucks eta using regression with an R2 score of 0.94. Created reports in order to monitor trip-trace data quality, data loss and eta engine accuracy.
• Analysed load variability in the mid mile network for Two month period. Carried out an in depth analysis of the utilisation of mid mile vehicles and highlighted the errors in weight and volume measurement for large shipments.

May 2018 – Jun 2018 · 1 mo · Gurugram

• Extracted, pre-processed and visualized borrowers data in order to understand the distribution of the variables.
• Created a classification model to predict whether a registered borrower will go live on the lending platform.
• Applied Logistic regression, decision trees and random forest on the data to get the best accuracy of 75% on the model.

May 2017 – Jun 2017 · 1 mo · Bengaluru Area, India

• Developed a Payment Reconciliation tool for Amazon and Flipkart portals.
• Created an operations manual for warehouse in a team of three members.
• Formulated a common payment template across different e-commerce sellers.
• Researched on Global supply chain companies in retail and fashion space like Zara,Amazon,7-Eleven etc