Agentic Data Engineering Fundamentals in 2 Hours
Published by O'Reilly Media, Inc.
From builder to strategist
What you’ll learn and how you can apply it
- Define “agentic” in practical operational terms and distinguish it from basic automation and GenAI copilot assistance
- Explain why fragmented data architectures prevent agentic AI from delivering business value
- Assess personal and organizational readiness for agentic transformation using a structured self-assessment framework
- Describe the strategic skill shift from code-first execution to strategy-first orchestration
- Quantify the “manual tax” by mapping time spent against tasks that AI-augmented workflows already handle in leading organizations
- Apply a decision matrix to categorize current work as automate, augment, or elevate and identify which category unlocks the most business value
- Build a stakeholder communication strategy that positions data engineering contributions in terms of business outcomes
- Design a 90-day upskilling plan that targets the strategic capabilities that agentic AI environments reward
Course description
With the shift to agentic AI, autonomous systems now execute the tasks that data engineers once built manually. The shift doesn’t eliminate the data engineer, but it does redefine the role. In this two-part course, Adam Morton gives practicing data engineers a concrete framework for navigating that redefinition.
In part I, you’ll learn what agentic AI is, why it’s changing data engineering, and what skills data engineers need to develop to work with it. Part II delivers the practical playbook—you’ll learn how to identify the “manual tax” that drains time and career capital, how to audit current work against an automate-augment-elevate decision matrix, and how to build the business literacy that positions data engineers as architects of autonomous systems rather than assemblers of pipelines. You’ll also examine real-world patterns and design principles from organizations that have already implemented unified, AI-driven data operations.
This live event is for you because...
- You’re a practicing data engineer or senior data analyst who wants to understand how agentic AI changes your work and how to position yourself for that change.
- You’re a data team lead, engineering manager, or data architect who’s rethinking how to structure and develop talent for roles in AI-assisted automation.
Prerequisites
- Hands-on experience building and maintaining data pipelines using SQL and Python
- Familiarity with at least one modern orchestration or transformation tool (Airflow, dbt, Prefect, Dagster, or equivalent)
- Basic understanding of one or more cloud data platforms such as Snowflake, Databricks, BigQuery, or Redshift
- No prior experience with agentic AI or LLM tooling required
Recommended preparation:
- Read Unlock Data Agility with Composable Data Architecture (book)
- Read Deploying AI Agents at Scale: 3 Patterns to Move Beyond POCs (book)
- Read “Data Strategies for AI Leaders” (MIT Technology Review Insights)
Recommended follow-up:
- Read Beyond Pipelines: Building Unified Foundations for AI at Scale (upcoming report)
- Read AI Engineering (book)
- Read Fundamentals of Data Engineering (book)
- Read Data Engineering Design Patterns (book)
Schedule
The time frames are only estimates and may vary according to how the class is progressing.
Part I: Why Your Role Is Evolving
What “agentic” actually means (20 minutes)
- Presentation: Defining agentic AI in operational terms—autonomous execution, dynamic adaptation, and the difference between copilot assistance and true agentic workflows; the spectrum—manual pipeline building, AI copilots, agentic systems; real-world examples—supply chain disruption response, automated data quality remediation, self-healing pipelines
- Group discussion: Where does your current stack sit on the agentic maturity spectrum?
Why fragmentation blocks agentic AI (15 minutes)
- Presentation: The hidden cost of siloed data architectures; why agentic systems require unified, trusted data foundations; the “silo tax,” or what fragmentation costs in latency, reliability, and agent decision quality; the DevOps parallel—how data/integration team unification mirrors the DevOps transformation of 15 years ago
- Q&A
The strategic skill shift (15 minutes)
- Presentation: From code-first execution to strategy-first orchestration; the new data engineer skillset—architecture thinking, business literacy, governance design; why this elevates engineers rather than replacing them; what the leading-edge looks like
- Group discussion: Where does your current role sit and where do you want it to go?
Agentic readiness self-assessment (10 minutes)
- Presentation: Introducing the readiness framework
- Hands-on exercise: Complete a quick maturity assessment against five dimensions—data foundation, team structure, governance, tooling, and skills
- Group discussion: Patterns and outliers
- Break
Part II: From Builder to Strategist—The Practical Shift to Strategic Thinking
Quantifying the manual tax (20 minutes)
- Presentation: What the manual tax is and what it costs; categories of work data engineers do today that agentic systems already handle in leading organizations; time audit methodology—mapping actual work against the automate-augment-elevate matrix; real examples of pipeline monitoring, schema drift detection, data quality alerting
- Hands-on exercise: Complete a quick personal time audit using a provided worksheet
The automate-augment-elevate decision matrix (15 minutes)
- Presentation: How to categorize work across three dimensions; automate (repetitive, rule-based execution with deterministic outcomes); augment (human judgment enhanced by AI-generated options, summaries, or anomaly flags); elevate (strategic, architectural, and cross-functional work that AI cannot replace)
- Demonstration: Applying the matrix to three common data engineering scenarios
- Q&A
Building business literacy and stakeholder influence (12 minutes)
- Presentation: How data engineers communicate value in an agentic world; translating technical contributions into business outcomes; working with cross-functional partners; data engineers as orchestrators of unified capability; the shift from reporting to stakeholders to designing solutions with them
- Group discussion: What business context do you currently lack and how would closing that gap change your influence?
90-day upskilling plan workshop (10 minutes)
- Presentation: The upskilling framework—three targeted skill tracks (architecture thinking, business literacy, governance design)
- Hands-on exercise: Draft a personal 90-day plan using a provided template
- Group discussion: What specific commitment are you making?
Wrap-up and Q&A (3 minutes)
Your Instructor
Adam Morton
Adam Morton is a data architecture author and thought leader who helps enterprise organizations navigate the transition from fragmented data systems to unified foundations capable of supporting agentic AI at scale. He’s the author of Unlock Data Agility with Composable Data Architecture (O’Reilly) and is completing Beyond Pipelines: Building Unified Foundations for AI at Scale, an O’Reilly report that examines how organizational design and technical architecture must evolve together for agentic AI to deliver business value.
Adam’s work focuses on practical, outcome-driven frameworks that senior practitioners and technical leaders can apply immediately. He regularly collaborates with enterprise vendors and research organizations including O'Reilly, Boomi, and MIT Technology Review Insights.