* Delivered data analysis and business intelligence projects using advanced features on Power BI, automated ETL pipelines on SQL Server, and coded forecasting models in MS Excel and Python, leading to an increase of revenue by 10% in 1 year
* Led the distribution chain of over 100 customers by providing ad-hoc data plus on time in full analyses to ensure customer success
* Automated 70% of manual business operations, reduced internal data request wait time by 90%, and minimized major risks by 40%
* Analyzed shipping costs and convinced the company to change to another shipper resulting in over $1M in savings for the fiscal year
* Developed dashboards and directly reported to the CEO and the COO of the company over 80% of the times. This increased data visibility resulting in effective communication between the senior leaders that drove the operational strategy across the company
* Leveraged analytic problem-solving skills in working with large disparate datasets to report portfolio operational metrics and performance. This ensured stakeholder and management visibility into operations and key performance indicators
* Managed and analyzed weekly, monthly, and quarterly costs using the external data feed by partners like Amazon, FedEx, UPS, etc.,
* Built a predictive model in Python for ‘Customer’s next purchase day’ with an output accuracy of 84%. Developed dashboards using DAX and Power Query on Power BI and communicated results to the sales and marketing teams for planning of future sales goals
* Acted as a liaison between the customers, operations, finance, and technology teams to ensure overall operational success
* Analyzed, classified, studied 6000 user comments on Reddit and developed a social media analytics perspective model through text analytics using deep learning LSTM (Long Short-Term Memory) classifier
* Leveraged Keras library in Python for sentiment analysis to augment human interactions and achieved an accuracy score of 73%
* Amplified marketing impact and used Tableau to visualize data points and communicated results and key ideas to decision makers
* Improved the business sales by 15% by applying data modeling and machine learning methods in Python to predict delinquent lottery retailer accounts for the sales & marketing team. Used SQL, Excel to integrate data from multiple sources
* Developed a predictive analytical model in Python to find the “Recency, Frequency, and Monetary (RFM) values” of customers
* Collaborated with multi-functional teams to facilitate new business models and communicated ideas to executives across the company
* Designed SQL queries to extract data from the data warehouse to the MySQL cluster by ensuring 100% dataflow accuracy to clients
* Created data reports and provided recommendations for specific scenarios which reduced validation effort by 20%
Assuming how a software company works, I developed an Entity Relationship Diagram and a Database Schema to show the workflow starting from the managers to the projects deployed to their clients/end-users.
Technologies: MySQL workbench
Concepts: Enterprise Data Modeling, Database Design
Visualized Los Angeles Crime Dataset of 2012-2016 and created a Dashboard for the same showing various insights and trends.
Technologies: Tableau Desktop
Concepts: Data Visualization
Used Python3 to train the best classification models for Santander’s customer base, predict potential churners, create visualization and spot key elements to improve retention rate.
Technologies: Python3
Concepts: Data Mining/Machine Learning, Predictive Analytics, Business Intelligence
Used Python3 to train the best classification models for a Portuguese Bank, prediciting the reach of their target market
Technologies: Python3
Concepts: Data Mining/Machine Learning, Predictive Analytics
Used Python3 to predict the probability that a customer would buy a quoted insurance plan by making use of various classifiers. Hypertuned their parameters to obtain the best scores, performed ensembled stacking to obtain the best classifier
Technologies: Python3
Concepts: Data Mining/Machine Learning, Predictive Analytics
Created small clusters using K-means and manually hypertuned the parameters and obtained the best set of clusters. Validated similarity of time-series by making use of matplotlib library in Python
Technologies: Python3
Concepts: Data Mining/Machine Learning, Clustering
Solved a semi-structured Machine Learning problem by making use of Python's Natural Language Tool-kit. Performed Wrapper and Filter type methods on different classifiers and obtained the best features of the document.
Technologies: Python3
Concepts: Data Mining/Machine Learning
By using the Keras neural network library, I performed tuning of the CNN model by experimenting the CNN parameters and drop-out rates and enchanced the accuracy from 65% to 86% of CIFAR-10 Image recognition problem.
Technologies: Python3 - Keras
Concepts: Machine Learing, Convolutional Neural Networks
By taking a huge data-set from kaggle, I used Stat-tools to perform Logistic regression for predicting the graduate admissions from an Indian perspective
Technologies: MS Excel, Stat-tools
Concepts: Logistic Regression
Designed a Lego race car using a full factorial design of experiments with 2 replicates on 4 selected variables to optimize the setting to obtain maximum distance travelled from a ramp. Financial Analysis and model adequacy checks, including residual analysis was done on minitab
Technologies: Minitab
Concepts: Design of Experiments
Designed and Developed a web application that displays the list and information of the top 10 movies of Indian Cinema.
Technologies: HTML5, CSS
Concepts: Web Technology
Developed a Stop-Wait protocol in Java
Technologies: Java (eclipse)
Concepts: Computer Networks
* Educated public, current, and prospective students across the United States about the Master’s in Business Analytics program
* Increased student engagement rate by 10% by conducting campus activities (seminars and tours) for them, their families, and agents
* Revamped the Business Analytics course structure and content, increasing student satisfaction rates from 30% to 65%
* Served as a liaison between recruiters and students which resulted in a 10% increase in applications. Stored and filtered out students based on their qualifications when they applied for jobs.
* Stored and filtered out students on Excel based on their qualifications when they applied for jobs.
* Organized various social community events which educated students on various sectors