Data Engineering with Generative and Agentic AI on AWS: Building an AI-Augmented Data Practice for the Enterprise

Enterprise data work now spans far more than moving records from one system to another. In Data Engineering with Generative and Agentic AI on AWS, Justin J. Leto explores how modern data platforms, cloud services, and AI-driven workflows fit together—from governance and architecture to retrieval, analytics, and agents.

The book balances technical design with the business context behind it. Readers encounter AWS-focused approaches to building and operating data systems, alongside questions of security, strategy, and how AI is changing the work of data teams.

Start with the foundations of data practice

The opening chapters introduce data engineering, the business value of data, and concepts such as distributed computing and the CAP theorem. They also frame generative and agentic AI within a modern data strategy, giving readers context for the practical architecture discussions that follow.

Design for governed, usable data

Security and governance receive substantial attention, including identity and access management, encryption, sensitive-data handling, compliance, auditing, and AI-specific risks such as prompt injection. The book then turns to data lakes, covering Apache Iceberg, S3 Tables, AWS Glue, Lake Formation, and the medallion architecture. These topics connect platform choices with the everyday needs of access control, data organization, and maintenance.

Connect teams, pipelines, and AI services

Further chapters address data mesh design with Amazon DataZone, data processing and transformation with AWS Glue, pipeline orchestration and observability, and multimodal data extraction. The coverage extends to retrieval-augmented generation, vector databases, streaming workloads, Redshift and text-to-SQL reporting, and generative business intelligence.

Explore agentic patterns on AWS

The final technical chapters focus on building AI agents with Bedrock AgentCore, Strands Agents, and the Model Context Protocol (MCP). Across the book, the emphasis is on seeing these tools in relation to the broader data engineering lifecycle—not as isolated demonstrations.

For data and technology practitioners

Data engineers, analysts, architects, and technology leaders interested in AWS data platforms and AI-enabled workflows will find a wide-ranging view of the field. The mix of architecture, implementation topics, governance, and business strategy makes this a useful choice for readers considering how enterprise data practices may adapt as generative and agentic AI become part of the toolkit.