I am experienced in leveraging agile frameworks to provide a robust synopsis for high level overviews. Iterative approaches to corporate strategy foster collaborative thinking to further the overall value proposition.
Developed an automated ticketing system using SAP CRM and ABAP for 15 over SAP departments. The system extracts information from customer email, creates a service request, and sends an acknowledgment response. Developed rapid application tools on SAP WebUI, that monitor ticket influx, KPIs and user data, and help improve resolution efficiency by 67%. Optimized complex SQL queries that pull data from large datasets with over 18 million records on a bi-monthly basis. Reduced query runtime by 27% per query.
Developed an appointment tool for TCS employees using AngularJS and Java Spring Hibernate ORM integration. The tool is used by over 300,000 employees for scheduling internal meetings and webinars.
GPA: 3.53/4.00
GPA: 7.8/10.0
Executed spatial statistics to the New York taxi dataset and identify the fifty most significant hot zones within the city. Developed and executed range join query, distance query, hot-zone analysis, and hot-cell analysis using Scala. Implemented Hadoop Distributed File System as a distributed storage system and Native Spark as the cluster manager.
Implemented the Bidirectional Search (BDS) Algorithm with a variation of manhattan heuristic for Pacman Domain. Performed a comparative performance analysis of BDS with other algorithms such as Breadth-First Search, Depth-First Search, A-Star Search over nineteen different layouts in the Pacman world, and also diagnosed outliers.
Developed an Android app that recognizes users’ food items and adds related calories to the calorie tracker. Integrated the MobileNet model, a Tensorflow Lite API for CNN based image classification to detect the food items. Proposed the crowdsensing model to expand the app to be able to recognize home-cooked food items.
Extracted skeletal key points from five-second videos depicting American Sign Language using Tensorflow PoseNet. Performed feature extraction and implemented machine learning classifiers such as Random Forest, Multi-Layer Perceptron, and KNN on the skeletal key points to predict the ASL gesture and compared the accuracy of the models. Developed a RESTful Flask application for the ML framework and deployed it as a cloud instance using AWS EC2.
Developed feature extraction algorithm using Matlab to identify different activities using raw sensor data collected from MyroBand. Analyzed the pattern of data streams in user dependent analysis using decision trees, neural networks, and SVM; and independent manner to derive a clear distinction between actions.
Designed a 3-layer neural network to compare various optimization algorithms on Fashion MNIST dataset. Implemented RMSProp, ADAM, AdaGrad, AdaDelta, Polyak's, and No-momentum and compared optimizers for convergence and stability using Python, Numpy and Pandas.
|
|