Mayank Vir Email & Phone Number

New York City Metropolitan Area

• Utilized techniques like Kaplan-Meier analysis, Cox proportional hazards regression modeling as well as Monte Carlo simulation.
• Developed study design utilizing targeted maximum likelihood estimation (TMLE) and G-computation.
• Implemented Multivariate Imputation by Chained Equations (MICE) method to handle missing covariates.
• Engineered ETL pipelines in Databricks handling SDTM, ADaM, and other diverse data formats.
• Automated descriptive analysis and reporting, saving at least 20 hours/week.
• Developed impactful data visualization dashboards to drive insights and decision-making.

Rochester, New York, United States

• Led data ingestion initiatives, managing storage and retrieval processes for a high-volume data lake hosted on AWS S3.
• Designed and maintained data catalog tables on Amazon Athena, enabling structured data querying and analysis for stakeholders.
• Conducted end-to-end unit testing to validate data pipelines, guaranteeing data accuracy and reliability.
• Developed real-time dashboard for marketing team using Twitter API, Pandas, Matplotlib and Tableau.
• Created telegram bot to collect inventory data from multiple sources via web scraping and APIs.
• Applied NLP and ML to build parser for ingesting varied inventory sources.

Delhi, India

• Developed Python scripts leveraging AWS S3, Lambda, and Athena to build cloud data platform ingesting retail transaction data.
• Processed data into Parquet format and stored securely in S3 data lake architecture optimized for analytics.
• Performed data transformations using Athena, SQL and Python lambda functions for cleansing and enrichment to enable analysis.
• Built QuickSight dashboards connected to S3 data lake to provide self-service visualization of customer purchasing behavior.

Bengaluru Area, India

• Project: Call Log Analysis (Banking Infrastructure Analysis, Client: Morgan Stanley).
• Analysed call logs addressing fraud detection, network health, audio data trends for a top US bank.
• Managed large volumes of data from log processing systems (SPLUNK), Kafka, Spark streaming, and PostgreSQL.
• Optimized SQL queries for faster OLAP and OLTP in PostgreSQL, utilizing concepts like shredding and column stores.
• Project: Fraud Analytics - Real-Time Decision (Client: Morgan Stanley).
• Implemented a real-time fraud prediction system, integrating ML models with existing fraud policies.

Master'S Degree, Data Science And Applications, A

Education record