AMAR JEET KUSHWAHA

● Over 4 years experience in Data Science and Machine Learning Statistical Analysis, Predictive Modeling, Python.

 ● Developed data-driven insights & solutions with advanced analytics (machine learning, deep learning and statistical models) to extract insights from large proprietary data sources Experience with common Data Science toolkits, such as Python, SQL, and QuickSight.

 ● Competent in analyzing the statistical data, & recording appropriate market behavior over a defined period of time to generate valuable insights using collected data for strategizing and forecasting future business plans

● Developed and implemented RFM (Recency, Frequency, Monetary) and product propensity models, leading to improved conversion rates and effective customer segmentation for targeted marketing campaigns. 

● Successfully implemented a Fraud Rule Engine, resulting in a savings of Rs. 32L in cashback within a span of 2 months by identifying and preventing fraudulent users on the platform. ○ Credit risk assessment and prevention. ○ Identifying creditworthy users on the platform using alternative(device) data. ○ Early identification of default users. 

● Developed propensity models for various products, including Virtual Credit Card (VCC), Buy Now Pay Later (BNPL) with an 11% increase in approval rates, and Rentals with a 2% increase in conversion rates. 

● Scoring mechanism for customers to prevent fraud in (VCC and BNPL). 

● Led initiatives in cashback optimization for customer retention, utilizing advanced models and rule engines to optimize customer retention costs from Rs. 34 to Rs. 21 , resulting in increased profitability and improved customer loyalty. 

● QuickSight DashBoard (Build Several Dashboards). 

● Proficient in geospatial data mapping, including reverse geocoding (converting lat-long coordinates to pincode), enabling accurate location-based analyses. ○ Geo mapping boosts our rental model by 30% MOM in the beginning of the project, helps to identify the right customer. 

● Skilled in performing similar user clustering and targeted marketing for rental products, leveraging geospatial data and customer segmentation techniques. 

● Build a Geo Reverse Encoding system an alternative to google map API. ○ Can map 3cr lat-long to pincode in 11 min. 

● Collaborate cross-functionally with teams such as marketing, finance, and engineering, ensuring seamless integration of analytics insights into product development and marketing strategies. 

● Utilized diverse data sources such as appographic data, SMS data, and visit data to augment traditional transactional data, enabling the creation of a comprehensive 360-degree view of the user base.

● Automated data processing pipeline. 

● Acquiring data from primary or secondary data sources and analyzing data by using statistical techniques and providing ongoing reports. 

● Sanity check tool(sync data between oracle and athena). 

● Developed Data monitoring tools (Audi, Mercedes-Benz, VW). 

● Data scraping tools saved the 9hr of the testing team. 

● Shop location mapping by google map scrapping , google search scraping, natural language processing. 

● Develop the customer segmentation tool for Audi. 

● Bugzilla Monitoring tool(prepare the report to see the performance of employees).

B.tech in computer science