End-to-End Walmart Retail Sales Analytics Platform with Databricks & Delta Lake

• Designed and implemented an end-to-end Retail Sales Analytics Platform using Databricks and Delta Lake, following the Medallion Architecture (Bronze, Silver, Gold) to ensure scalable and maintainable data processing.
• Ingested raw Walmart retail sales data (CSV format) into a Bronze layer as Delta files, preserving source fidelity and adding ingestion metadata to support traceability and reprocessing.
• Built Silver layer transformations using PySpark to clean and standardize data, handling real-world issues such as malformed dates, inconsistent numeric values, and duplicate records.
• Modeled cleaned data into Orders, Customers, and Products tables, applying schema enforcement and safe casting techniques (try_cast, try_to_date) to prevent pipeline failures.
• Designed an analytics-ready Gold layer using a star schema, creating fact and dimension views to support business intelligence and ad-hoc SQL analytics.
• Applied Unity Catalog governance to manage schemas, enforce table registration, and ensure controlled access to curated datasets.
• Implemented performance optimizations including Z-Ordering, table statistics, and data quality constraints to improve query performance and ensure reliable analytics.
• Delivered BI-ready datasets optimized for reporting, trend analysis, and KPI computation, aligned with enterprise-grade data engineering best practices.