Description
This issue proposes the development and integration of AI-powered features to assist users throughout the data ingestion process in the D2E platform. The goal is to streamline and automate the key stages of data onboarding—such as concept mapping (e.g., mapping local codes to OMOP concepts), structural mapping (e.g., aligning source data fields with OMOP CDM tables), and ETL pipeline implementation (e.g., generating ETL code from mappings or descriptions)—reducing manual workload and increasing accuracy, transparency, and reproducibility.
The AI-driven enhancements should support all these areas, making the data ingestion process more efficient, reliable, and accessible, and empowering users to onboard new data sources with greater speed and confidence.