Akshay Kumar

Experienced professional with strong background in social media strategy, content development, and brand promotion. Demonstrated expertise in event management, inventory optimization, and customer relations. Skilled at driving sales, contributing to strategic planning, and leading teams to project success. Committed to delivering exceptional results through effective communication and proactive approach.

• IIM- Ahmedabad - 1st prize- Group dance
• FMS- Delhi University - 1st prize Group dance
• SRCC- Delhi University - 1st prize - Group dance
• Head Volunteer at Art of Living Maho-utsav

• Channel V India Fest Goa (2014)
• Entrepreneurship Convention 1.0
• Football Analytics using Python

• Project : 1 -
• Optimizing Insurance Conversion Rates through AI and ML - Worked on an AI-driven insurance sales lead conversion prediction model to analyze the probabilities of conversion likelihood. Processed data and implemented Recursive Feature Elimination (RFE) to identify the top 15 attributes. Employed logistic regression to assess p-values, eliminating columns with p-values significantly different from
• Achieved an accuracy of 89.56% with a 0.5 probability threshold. Created an ROC curve to evaluate sensitivity versus specificity and identified a cut-off point of 0.2 for optimal performance. Computed precision, recall, and accuracy, resulting in an increase to 90.34%. Subsequently, explored other models and selected 'Random Forest.' Initially, without defining the depth, the accuracy was 89.97%. However, after setting the depth to 5, the accuracy improved to 91.14%.,
• There is a substantial opportunity to apply Lead Scoring beyond the insurance sector, such as in Banking and Education, where there are high volumes of leads. On average, there's nearly a 50% lead decay due to extended time before initial contact. This represents a significant opportunity cost that could be salvaged through effective lead scoring strategies.
• Project 2-Time Series Analysis: Forecasting Store and Product Family Sales Trends-
• Led a comprehensive Time Series Analysis project focusing on sales prediction for 33 product families across 54 store combinations. Utilized datasets including Train, Test, Store, Transactions, Holidays and Events, and Daily Oil Price to develop robust forecasting models.
• Implemented a use case approach by performing essential tasks such as packages installation, dataset importing, and correlation analysis between 'Total Sales' and 'Transactions.' Established a strong correlation coefficient, demonstrating a significant relationship between revenue generated and individual sales transactions.
• Developed and executed comprehensive sales forecasting business cases across diverse industries, including retail, e-commerce, automobile dealerships, hospitality, and consumer goods. Leveraged advanced predictive modeling techniques such as Linear Regression and Gradient Boosting Machines (GBM) to optimize sales strategies, inventory management, and marketing campaigns for improved business outcomes.