Pratham Sharma
Data Scientist @Four Seasons
Signup · Get unlimited contacts
WORK HISTORY
Data Scientist @Four Seasons
Toronto, ON, CA
Key areas of focus: Time Series Forecasting, Recommendation Systems, Natural Language Processing, Machine Learning• Led end-to-end time series forecasting for digital revenue and booking performance across web andmobile, enabling Digital and Commercial teams to set monthly targets and guide planning decisions• Improved forecast accuracy and explainability by integrating 5+ data sources and introducing a formal model evaluation framework• Took primary ownership of the forecasting workstream with minimal oversight, building strong cross-functional trust with partner teams and stakeholders• Built and productionized recommendation models for a strategic guest experience initiative, embedding business rules to optimize recommendation relevance and guest engagement• Operationalized recommendation models using MLflow and Azure Data Factory for automated weeklyinference, enabling scalable and repeatable deployment• Analyzed large-scale call transcript data and developed NLP models for topic classification and issueresolution tracking; identified key data quality gaps and partnered with vendors to improve data readiness
EDUCATION
Western University
Master's degree, Data Analytics
Navrachana International School
CBSE Higher Secondary, Science - Mathematics
Vellore Institute of Technology
Bachelor of Technology - BTech, Computer Science and Engineering
ABOUT PRATHAM SHARMA
A data science professional with expertise in Advanced Analytics, Predictive Modelling, and Business Intelligence, I help organizations make data-driven decisions, solve problems, and generate value by harnessing the power of Data Science and Advanced Analytics.Professional Experience:3 years of experience as a Data Scientist across the Hospitality, Healthcare, and Financial Services sectors, building data-driven solutions that help organizations tackle crucial business problems, improve operational efficiency, and make data-driven decisions.Key Achievements:• Developed and deployed forecasting models to forecast emergency department visits with 4% MAPE, leading to a 12% reduction in ED wait times.• Led the development of multiple Power BI dashboards for cross-functional teams to facilitate informed decision-making.• Leveraged statistical modeling techniques to implement algorithms that can accurately detect anomalies in financial transactions, resulting in a significant 56% decrease in the time taken to detect fraudulent activity.Areas of expertise:• Building predictive analytics and forecasting models using machine learning and time series analysis techniques• Performing data analysis and statistical modeling to uncover patterns and identify actionable insights• Data cleaning, data modeling and data visualization to convey key insights to stakeholders• Effective communication of results and details of technical solutions to stakeholders to facilitate informed decision-making• Collaboration with cross-functional teams to understand business requirements and propose suitable technical solutionsCore Skills:• Data Science Tech Stack: Python, PySpark, R, SQL, Azure Databricks, Microsoft Azure, Power BI• Frameworks & Libraries: Pandas, NumPy, Scikit-Learn, Matplotlib, Seaborn, TensorFlow, MLflow• Techniques & Concepts: Machine Learning, Time Series Analysis, Natural Language Processing, Recommendation Systems, Statistical Modeling, Hypothesis Testing, Data Warehousing, Data Pipelines• Other Tools: Git, Microsoft Fabric, Tableau, Azure Machine Learning, Azure Data Lake, Azure Data Factory, Apache Airflow• Project Management: Git, GitHub, Azure DevOps, JIRAI’m always eager to connect with fellow professionals, recruiters, and organizations looking to leverage data science and advanced analytics. If you are interested in discussing potential opportunities or collaborations, let’s connect and explore how my experience and skills can add value to your organization.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.