VEDIKA PUSHKAR BHENDE

Successful team leader and complex problem-solver skilled in reviewing, fixing and improving code to maximize product performance and usability. Bringing 2+ years of software development and management experience.

• MIE Department, UMass Amherst: Engineering Management Program Student Ambassador (09/22 – 05/24)
• Sports Club, CCEW: Table-Tennis Sports Captain – won 11 All-India inter-college tournaments (06/17 – 07/20)
• Cultural Club, CCEW: Digital Media and Communications Lead (07/19 – 09/20)

• Project Management and Leadership (UMass Amherst)
• Business Analysis & Process Management
• Data Analytics for Lean Six Sigma (University of Amsterdam)

Southwest Airlines Airfare Prediction (11/2023 - 12/2023)

• Predicted the airfare of airline routes using various independent variables and explored four regression models for optimal prediction of airfare, evaluated using the lowest Mean Squared Error value, with log-linear regression giving the lowest error rate of 4.25%
• Implemented various methods – Regression using XLminer in MS Excel, Linear Regression using feature selection and Log-Linear Regression using R programming and used SAC to find the contribution of each variable to the model

Twitter Airline Sentiment Analysis (11/2022 - 12/2022)

• Applied Natural Language Processing to detect the sentiment of tweets targeting airline companies to classify them as positive or negative reviews
• Analyzed the performance of prediction accuracy using four analytical models – Support Vector Machines, Multinomial Naïve Bayes, Random Forest Classification and Logistic Regression, and evaluated the models using various statistical parameters, with 87% optimal test set accuracy

Unsupervised Learning Algorithm Design (10/2022 - 11/2022)

• Developed and implemented the K-means clustering algorithm from scratch, as a way of analyzing clustering techniques
• Conducted a comprehensive comparative analysis between the custom-developed algorithm and the scikit-learn implementation, and concluded that the performance of the custom-built algorithm is 5% superior, with lower error

Optimal Energy Portfolio Decision Optimization (10/2022 - 12/2022)

• Analyzed data of various energy portfolios to decide the best approach for optimizing energy policies to better fight climate change, and presented sets of optimal portfolios for the decision maker to select the best suited solution as per their preference

Video Recommendation System (corporate sponsor: kPoint Technologies) (01/2020 - 06/2020)

• Led a team that designed and implemented Machine Learning algorithms to recommend videos for a user based on past viewership data retrieved from the company database, with a prediction accuracy of 68%
• Developed a backend microsystem that is presently deployed on the company’s streaming platform for their global clients