What does that data pipeline do? What were its sources, transformations, and target? How did you use Snowflake for fact and dimension tables?

The pipeline I built was an end‑to‑end ETL that ingested sales and inventory data from on‑premise SQL Server and an external REST API, performed cleansing and enrichment, and loaded the results into Snowflake for reporting. The sources were a transactional database (SQL Server) and a JSON feed from a partner system. Transformations included deduplication, type conversion, date normalization, and joining the two sources on product SKU. I also calculated derived metrics such as inventory turnover and sales velocity. The target was a Snowflake data warehouse where I created a star schema: fact tables for sales and inventory, and dimension tables for products, customers, and time. Snowflake’s semi‑structured support allowed me to load the raw JSON directly into a variant column, then use SQL to flatten and transform it. I leveraged Snowflake’s clustering keys to optimize query performance on the fact tables. The pipeline was orchestrated with Azure Data Factory, which scheduled the ingestion, triggered Snowflake COPY commands, and monitored job status. This design ensured scalability, maintainability, and fast query response for business users.