Scalable Data Product Value Management with Agent ready SDK

Build AI data products with the ODPS family, SDK validation, product catalogs, value graphs, and agent-ready workflows.

New

Created byDr. Jarkko Moilanen

Last updated 6/2026

English

What you'll learn

Explain how ODPS, ODPC, ODPG, and ODPV work together as one standards family
Use the Open Data Products SDK to validate, inspect, and explain standards files
Build a simple data product value graph with products, use cases, business objectives, KPIs, and signals
Connect ODPS product definitions with ODPC catalog objects and ODPG graph relationships
Generate a basic value graph report that highlights product-to-value links and missing connections

Course content

5 sections • 23 lectures • 3h 40m total length

Course introduction and learning outcomes4:09
In this lesson, you will get a clear overview of the course structure and what you will build toward.
You will learn to:
• Understand the purpose of the ODPS family training course
• See how the lessons, examples, and resources fit together
• Prepare for the shift from specification knowledge to practical use
Includes: course overview and supporting resources.

Why ODPS became a family: The Challenge4:37
In this lesson, you will see why ODPS had to grow from one specification into a wider standards family.
You will learn to:
• Understand why one specification is no longer enough
• Explain the shift from single product metadata to a wider standards family
• Recognize the need for catalogs, graphs, vocabulary, and tooling around data products
Includes: lesson slides and the Open Data Product Standards homepage.
Why ODPS became a family: The Solution6:10
In this lesson, you will see how the ODPS family answers the limits of a single specification.
You will learn to:
• Understand the role of each standard in the ODPS family
• Explain how product specs, catalogs, graphs, and vocabulary work together
• Recognize how the family structure supports scale, interoperability, and reuse
Includes: lesson slides and the Open Data Product Standards homepage.
Why AI agents need structured data product context10:04
In this lesson, you will learn why AI agents need structured, standards-based context to work safely with data products.
You will learn to:
• Understand why unstructured documentation is not enough for AI agents
• Explain how standards give agents clear product context and boundaries
• Recognize why machine-readable metadata matters for future data product operations
Includes: lesson slides and supporting references.
How the ODPS family works together6:24
In this lesson, you will see how the ODPS family connects product specs, catalogs, graphs, and vocabulary into one operating model.
You will learn to:
• Understand how the ODPS family components support each other
• Explain the flow from individual data products to catalogs and connected graphs
• Recognize how tooling and standards help teams manage data products at scale
Includes: lesson slides and supporting references.
Quiz 1 Understanding of the ODPS family foundation

What the Open Data Products SDK is11:33
In this lesson, you will understand what the Open Data Products SDK does and why it matters.
You will learn to:
• Explain the role of the SDK in the ODPS family
• Understand how SDK commands support validation, explanation, generation, catalogs, and graphs
• Recognize why tooling makes standards easier to adopt in real workflows
Includes: lesson slides and SDK references.
Installing and running the SDK4:42
In this lesson, you will prepare your environment and run the SDK for the first time.
You will learn to:
• Install the Open Data Products SDK
• Run basic SDK commands from the command line
• Confirm that your setup is ready for the practical exercises

NOTE! We use Google Colab in the exercises.
Materials you will find useful:
Download the attached Colab file. You need to open that resource in Google Colab!
You can also download the Installation guide. It contains how to install the SDK on local machine on terminal.
In case you dont use Github, I also added html.zip containing HTML version of the exercise materials and commands.
Validating and explaining standards files8:44
In this lesson, you will use the SDK to validate and explain standards-based files.
You will learn to:
• Validate ODPS family files against their expected structure
• Read validation feedback and understand what needs correction
• Use explanation commands to make standards files easier to understand
Includes: practical SDK examples and supporting files.
Use the ODPV Vocabulary Helpers12:05
In this lesson, you will work with ODPV vocabulary helpers in the SDK.
You will learn to:
• Understand why controlled vocabulary matters in data product metadata
• Use SDK helpers to inspect and apply vocabulary values
• Improve consistency across product descriptions, catalogs, and connected graphs
Includes: vocabulary examples and SDK commands.
Working with local LLMs13:13
In this lesson, you will see how the SDK works with local language models.
You will learn to:
• Understand when local LLMs are useful for data product work
• Connect the SDK generation workflow to a local model setup
• Recognize the tradeoffs between privacy, control, performance, and model quality
Includes: setup guidance and local generation examples.
Working with online LLM providers12:38
In this lesson, you will see how the SDK works with online LLM providers.
You will learn to:
• Understand how provider presets simplify LLM configuration
• Use the same generation pattern with online and local models
• Recognize how provider settings make the SDK easier to extend
Includes: provider examples and generation configuration guidance.
Generating ODPS Data Products from business requirements14:56
In this lesson, you will turn human-written source material into structured data product content.
You will learn to:
• Understand how business requirements become data product intent
• Use the SDK to generate ODPS product specifications from source documents
• Recognize how TOON and GCF sidecars support generation workflows
Includes: source material examples and SDK generation guidance.
Working with Generation and Fragments15:54
In this lesson, you will understand how generation outputs and fragments support catalog and graph workflows.
You will learn to:
• Explain the difference between a product specification and a product reference fragment
• Understand how fragments support catalogs and connected graphs
• Use generated fragments as reusable building blocks in later workflows
Includes: generation examples and fragment output files.
Working with ODPC Catalogs7:13
In this lesson, you will see how ODPC catalogs organize data product references.
You will learn to:
• Understand the role of catalogs in the ODPS family
• Use product reference fragments inside a catalog context
• Recognize how catalogs help teams manage collections of data products
Includes: catalog examples and SDK workflow guidance.
From Catalogs to Connected Graphs17:50
In this lesson, you will move from cataloged data products to connected data product graphs.
You will learn to:
• Understand how catalogs provide structure for connected graph generation
• Explain how ODPG connects products, signals, roles, use cases, and relationships
• Recognize how graph views help reveal the wider data product ecosystem
Section 3 recap, From SDK commands to practical workflows8:10
In this lesson, you will review how the SDK capabilities connect into practical workflows.
You will learn to:
• Summarize the main SDK commands covered in this section
• Explain how validation, vocabulary, generation, catalogs, and graphs fit together
• Prepare for the practical workflow exercises in the next section
Includes: section recap slides and workflow summary.
Quiz 2 SDK foundation and generation workflows

From Commands to Portfolio Builder4:08
In this lesson, you will see how separate SDK commands come together into one portfolio-building workflow.
You will learn to:
• Understand why the portfolio builder is the final practical stage
• Explain how validation, catalogs, graphs, and HTML output connect
• Recognize how the SDK turns standards files into a reviewable portfolio
Includes: lesson slides and workflow guidance.
What the Portfolio Builder Creates6:11
In this lesson, you will understand the outputs created by the portfolio builder.
You will learn to:
• Identify the main files and folders created by the portfolio workflow
• Understand how catalog, graph, and HTML outputs support different users
• Recognize the difference between human-friendly views and machine-readable files
Includes: output examples and supporting materials.
Build the First Portfolio in Colab14:55
In this lesson, you will build your first data product portfolio in Colab. Download from resources and open in Colab
You will learn to:
• Run the portfolio builder in a guided Colab environment
• Generate a working portfolio from prepared standards files
• Review the first portfolio output as a practical result of the course
Includes: Colab workbook and example files.
Update the Portfolio and Review Version History9:36
In this lesson, you will update the portfolio and see how changes appear over time.
You will learn to:
• Add new content to an existing portfolio
• Rebuild the portfolio after an update
• Review version history to understand how the portfolio changes
Localize the Portfolio Catalog and Graph12:02
In this lesson, you will localize portfolio content for multilingual data product use.
You will learn to:
• Understand how localization supports wider data product adoption
• Apply localized metadata to catalog and graph outputs
• Review how multilingual content improves usability for different audiences
Final Review: Human View and Agent-Ready YAML13:17
In this lesson, you will review the final portfolio from both human and machine perspectives.
You will learn to:
• Inspect the human-friendly HTML portfolio view
• Review the YAML files that make the portfolio structured and reusable
• Understand why the same portfolio must serve people, platforms, and AI agents
Quiz 3 Portfolio-building workflo

Requirements

Basic understanding of data products, metadata, data catalogs, or data governance is helpful, but not required
Basic Python knowledge is useful for the SDK and practical build sections
A computer with Python installed is recommended for following the hands-on exercises
No prior experience with ODPS, ODPC, ODPG, ODPV, or the Open Data Products SDK is required

Description

Data products are entering a new phase.

AI agents, copilots, and automation tools need more than documents and dashboards. They need clear data product context. They need product definitions, catalogs, links, shared words, checks, and machine-readable files.

This course shows how the Open Data Product Standards family helps you build that foundation.

You will learn how ODPS defines a data product. You will see how ODPC organizes data products into catalogs. You will learn how ODPG connects products, use cases, goals, KPIs, signals, and roles into value graphs. You will also see how ODPV keeps words and meanings consistent.

The course has 23 lessons and 2 practical parts. Around 50% of the course is hands-on SDK work.

You will use the Open Data Products SDK to check standards files, explain structured metadata, use vocabulary helpers, create product content from business needs, work with fragments, build catalogs, and create connected graphs.

In the final practical part, you will build a simple data product portfolio in Colab. The portfolio helps you review whether a data product is valid, included in a catalog, linked to use cases, connected to business goals, measured with KPIs, and ready for human review and AI-agent workflows.

This course is for data product owners, data architects, governance leads, platform teams, AI product teams, and developers.

By the end, you will understand the ODPS family. You will also have a working SDK-based workflow that you can reuse and extend in real data product work.

Who this course is for:

Data product owners, data architects, data governance professionals, platform teams, and developers who want to understand the ODPS standards family and use the SDK in practice
People working with data catalogs, data product portfolios, knowledge graphs, AI agents, or business value mapping
Learners who already know the basics of ODPS and want to see how the wider standards family works through a practical graph-based example

Scalable Data Product Value Management with Agent ready SDK

What you'll learn

Explore related topics

Course content

Introduction1 lecture • 4min

The (Open) Data Product Standards Family4 lectures • 27min

Introduction to the SDK11 lectures • 2hr 7min

Final Stage: Build the Portfolio6 lectures • 1hr

Time to celebrate1 lecture • 2min

Requirements

Description

Who this course is for: