Data Frameworks for Generative AI

Overview

Generative AI systems—large language models (LLMs), Retrieval-Augmented Generation (RAG), and agentic AI—demand modern data strategies to ensure accuracy and reliability. These technologies hinge on high-quality, well-governed data; without a robust framework, even advanced models risk generating flawed outputs.

This course explores how foundational data principles enable scalable, trustworthy generative AI solutions. We start by analyzing LLMs’ role in today’s AI applications, addressing limitations like hallucinations and outdated context.

Next, we examine how RAG enhances LLMs with retrieval mechanisms, and why agentic AI—enabling autonomous reasoning and decision-making—is the next frontier. Each evolution underscores the criticality of structured, governed data.

You’ll learn the core components of modern data strategy:

unified frameworks, effective management, and principled governance. We dissect how structured and unstructured data uniquely power AI systems and introduce pillars like accessibility, security, lineage, and scalability.

Through case studies, hands-on exercises, and expert-led discussions, you’ll gain practical insights into data taxonomies, classification, and real-world implementation. By the end of the course, you’ll master applying these strategies to GenAI projects, ensuring systems are built on reliable, enterprise-ready data foundations.

This course is for Data Scientists, Data Engineers, AI or GenAI Leaders, and any curious learner who wants to understand modern data strategies, especially data frameworks and their impact and applications. By the end of this course, you will be able to, explain the components of a modern data framework and its role in GenAI.

You will also be able to differentiate between structured and unstructured data in AI implementations and apply foundational data governance and management principles to support scalable GenAI solutions.

Syllabus

• Understanding Modern Data Strategy Fundamentals
In today’s rapidly evolving AI landscape, data is no longer just a byproduct—it's the fuel that powers intelligent systems. This module introduces the foundational role of data frameworks in modern data strategy, especially in the context of Generative AI applications. We begin by discussing how Large Language Models (LLMs), RAG (Retrieval-Augmented Generation), and Agentic AI systems rely on high-quality, well-governed data. You’ll explore the evolution of these technologies, their dependencies on structured and unstructured data, and how data strategy must evolve in parallel. The module also covers the core pillars of a modern data strategy—data frameworks, management, and governance—and explains their critical role in driving performance, compliance, and scalability in GenAI solutions. Through examples, case studies, and guided walkthroughs, you’ll learn how to design frameworks that support relevance, quality, and accountability, ensuring that AI systems are both powerful and responsible.
• Introduction to a Comprehensive Data Frameworks
As generative AI continues to evolve, the importance of well-structured data frameworks has become central to building scalable and ethical AI systems. In this module, we focus on designing comprehensive data frameworks that support the needs of modern AI systems, especially those that rely on both structured and unstructured data. You’ll explore the role of customized taxonomies in organizing data, and how these taxonomies enable consistent data classification and retrieval. We also examine how Responsible AI (RAI) principles influence data strategy and governance, ensuring that fairness, transparency, and accountability are built into the foundation. Through practical discussions and expert insights, you'll see how the components of a robust data framework—taxonomy design, ethical considerations, and governance practices—work together. Finally, we look ahead at emerging trends and evolving expectations in data frameworks to prepare for the future of GenAI deployment.

Taught by

Fractal Analytics and David Drummond

Subjects

Computer Science