Ab Initio ETL Pipeline Development

The aim is to design and implement a production-ready ETL pipeline in Ab Initio that reliably moves data from multiple sources—relational databases, external APIs, and periodic flat-file drops—into our analytics warehouse.

Here is the scope as I see it:
• Create end-to-end Ab Initio graphs that extract, transform, and load data on a daily schedule.
• Build source connectors for our Oracle and Postgres instances, the provided RESTful APIs, and a shared SFTP location where CSV files arrive.
• Apply the required business transformations, cleanse the data, and map it into our warehouse schema.
• Include robust error handling, restartability, and performance tuning.
• Package everything with clear runbooks and in-line graph documentation so the in-house team can support it once delivered.

A successful hand-off means the graphs run unattended, load completion metrics are logged, and validation queries prove row-level accuracy across all sources.