Stop Guessing: Data Modeling and Strategy for Trustworthy AI

The effectiveness of any project is bottlenecked by the quality and structure of its underlying data. This session moves beyond traditional data warehousing to explore the critical modeling and strategic adjustments required to create an "AI-Ready" data ecosystem.

Developers will learn:

• How to evolve data models (from relational to wide-table/graph) to satisfy the demands of feature engineering and model training.
• The strategic necessity of Data Governance and Metadata Management to ensure data quality, lineage, and compliance for responsible AI.
• Architectural best practices, including the role of the Feature Store in ensuring training-serving consistency and feature reuse.
• Tactics for identifying and mitigating data bias at the source to ensure ethical, fair, and robust model outcomes.
• This is a practical guide for preparing your data architecture to support scalable, high-performing, and trustworthy AI/ML initiatives.

Target Audience

• Data Developers and Engineers
• ML Engineers and Scientists
• Data Architects and Data Strategists
• Software Developers interested in building data-intensive AI applications

Level of expertise

(Assumes a basic understanding of data concepts (databases, ETL/ELT) and a familiarity with the terms AI/ML, but focuses on introducing new, advanced strategic and architectural principles.)