• Developed a machine learning model for “Ground Water Rehabilitation Scheduling” for Ontario Municipality to predict remaining useful life of a well to increase life of the well. I presented this model at the National Ground Water Association, Ground Water Week in Las Vegas in December 2022. A summary of work was published in Water Canada magazine (May/June 2023 edition)• Designed and deployed python scripts on an Azure virtual machine to automate traffic growth analysis for the Transportation Department which eliminates labor intensive complex calculation and creates a report with visual maps within seconds• Developed a Go-No Go tool using XGboost machine learning classification model with Flask web framework to help the organization make key project decisions• Performed Hot Spot GIS cluster analysis using ArcGIS to identify opportunities in advancing business • Working on developing Multivariate Multistep deep learning model to forecast chloride levels using chloride, water quality and climate data to understand road salt application and improve salt control procedure

• Automated data analysis by establishing a data gateway between a python script, Power BI, and SQL server to deliver statistics such as gross and net labor earnings, 12 months rolling average, and median values of proposals by department, and proposal future projection to President, Vice President, and CFO • Designed ETL workflow to analyze project data and applied statistical methods and machine learning algorithms to achieve better results

Course - Statistical Methods for Machine Learning II• Grading daily quizzes and assignments• Helping professor in designing projects and assignments• Conducting problem solving sessions

Course - Data Science II and III

Work-Study:• Worked on the development and analysis of sports performance on Tableau• This resulted in the development of concise reports that were provided to the management

Developed a fact verification solution (web app) called 'Sat AI' (Team of 4), which analyzes factual correctness, gathers source backed relevant facts and analyzes tone and semantics of the content. It uses novel Natural Language Processing comparison layers to determine factual correct information from a consensus of information gathered from credible sources. Qualified for the final round of NEST 2020, a competition by the University of Toronto's startup incubator (UofT Hatchery).

Course - Professional Ethics and Practice

• Developed strategies to reduce waste and optimize inventory level through data analysis• Leading the team in commissioning of LDPE (Low-Density Polyethylene) plant• In charge of planning for Mechanical team in LDPE plant• Data analysis for optimization of the plant inventory• Performed material handling, procurement, and inventory modeling using SAP to optimize inventory and improve service