Vaishnavi Shankar Devadig Email & Phone Number

Naperville, Illinois, United States

• Collaborated on the development of Medicare claims ingestion workflows under new CMS regulations, leveraging Databricks, and Spark cluster for the ingestion of 1 TB claim data using PySpark. Performed data transformations, converting raw data into Parquet. • Developed Azure Data Factory pipelines to orchestrate the entire data flow, ensuring seamless and compliant processing, storage, and reporting of dental and pharmacy claims. • Developed and optimized SQL Server stored procedures to ingest processed data, supporting business-critical functions.• Designed and implemented a Generative AI model using transformer-based architectures to generate contextual summaries of claims, integrating it with existing workflows to assist the business team in faster and more accurate decision-making.• Analyzed, validated, and processed over 1 million healthcare claims, ensuring adherence to CMS requirements and mitigating financial risks by implementing robust compliance mechanisms.

Chicago, Illinois, United States

• Developed a user-tag-specific social and context-aware recommender system under Prof. Yong Zheng (IITC) and designed a novel evaluation metric for the model (BERTopic, Large Language Models, Generative AI, Recbole, DeepFM).• Performed comparative analysis on traditional, graph, and transformer models for research on improving recommender systems.• Conducted in-depth analysis of CTA usage patterns by integrating data from multiple sources, leveraging advanced statistical models such as linear regression, lasso regression, and clustering to study ridership trends and crime rates.• Investigated the impact of COVID-19 on transit usage, employing predictive modeling techniques to quantify changes in ridership and inform policy recommendations.• Served as a teaching assistant for Data Preparation and Analysis (CSP571) course, mentoring over 150 students, conducting lectures, grading examinations, and guiding them on statistical analysis projects.

Houston, Texas, United States

• Spearheaded the collecting and preprocessing of extensive inventory datasets across multiple client locations, employing Python for efficient data handling and preparation (Data Collection, Data Preprocessing, Python).• Implemented Python-based regression models and time series forecasting to enhance inventory management, achieving a 12% efficiency improvement and cost reduction. Deployed ETL pipelines using AWS and utilized Tableau for advanced data visualization, alongside model deployment on Amazon SageMaker (AWS Glue, Athena, Amazon SageMaker).

Mangaluru, Karnataka, India

• Led a comprehensive customer data analysis for a state-owned telecom company across 50+ sub-divisions, uncovering churn and segmentation insights that directly informed strategic decisions and retention policies.• Utilized T-SQL, Pandas, and NumPy to manage and preprocess large, complex datasets, applying advanced machine learning models such as clustering and classification to forecast churn and categorize customers effectively.• Designed and deployed interactive Power BI dashboards, delivering actionable insights and visualizations that supported data-driven decisions for 4G network implementation and enhanced customer engagement.

Bengaluru, Karnataka, India

• Developed and executed sophisticated SQL queries for in-depth analysis of vSphere customer behavior and system metrics, ensuring data integrity and providing valuable insights.• Engineered and refined ETL pipelines using PySpark, with deployments managed through Git, achieving a 25% reduction in execution times for processing extensive datasets.• Designed interactive Tableau dashboards to effectively visualize customer analytics, facilitating key metric monitoring and supporting strategic, data-driven decision-making.• Proactively identified virtual machines at risk of performance degradation due to resource constraints. Configured and resolved issues for global clients, optimizing vSphere environments to maintain peak performance efficiency.

Master Of Science - Mas, Data Science

Bachelor Of Engineering - Be, Computer Science