Looking for a fresh, fun, and surprisingly entertaining way to connect with AI and agentics? This turnkey case study uses a flipped classroom model

Try your hand at the latest markdown challenges - rules and headings.

Have you been listening to my podcast, The Plain Text of AI, watching my YouTube videos on preparing data for AI, or checking out the free Python Coding book I blogged about a few months ago, and found yourself wondering: What on earth does all of this have to do with Tableau?

The answer is: Tableau Agent.

If you peeked at my day-to-day routine lately, you’d see a major transformation—my entire workday is now spent interacting with AI using Markdown.  And imagine my delight when I realized I'd accidentally stumbled into a solid foundation for Tableau Agent (or Agentic AI.) Turns out, learning Python, writing in Markdown, and figuring out the best practices for conversing with AI for my podcast make the ultimate toolkit for this exact moment.

To help you bridge the gap between traditional dashboard design and the new agentic world, my next major example on the site will be a dedicated Markdown Exercise. It will walk you through everything I’ve learned about structuring text and data so you can get the absolute most out of Tableau Agent (or any AI for that matter.) I hope my simple tips will make your journey fun and productive?

My second podcast episode is live!

After Episode 1 dropped, a friend reached out to me with a very simple, direct question: 'Why do they get it wrong?'

I was excited to share Episode 2 to answer that exact question! There's a reason the show logo is a child-like chalk drawing, and I hope you enjoy listening to that funny real-world example as much as I enjoyed writing it!

Today, I’m sharing the first of several podcast episodes dedicated to AI.

My interest in AI began when I started wondering: How do we actually prepare data for Tableau’s agentic AI?

Before diving straight into the deep technical details of data mapping, semantics, and data lineage in the upcoming "Shaping Data for AI" podcast, I wanted to step back. In these initial episodes, I’m sharing what I’ve learned about AI foundations so far.

Check out the latest Viz Whiz - Case 14

I spent considerable time researching, writing and recording a short podcast series called The Plain Text of AI.

Today I made the transcripts available through OER Commons as free educational material.

AI can be exciting, confusing, useful, frustrating, and occasionally hilarious. My hope is that these stories help people better understand what AI is, what it isn't, and why humans still belong in the loop.

If even one student, teacher, or curious learner finds it useful, I'll consider the project a success.

https://oercommons.org/courseware/lesson/146360

Check out the latest Viz Whiz - Case 13

In this step-by-step hands-on data forensic investigation, we dive into root-cause analysis using Tableau. See if you can spot data bias, bypass misleading vanity metrics, and learn the difference between correlation and causation.

Data doesn’t have a single author; it has a committee of perspectives for how a metric is collectively defined. This example leverages the Data Mapping Files to provide a logic review in real-time from the viewpoint of each persona involved, and demonstrates the clear deterministic intent outlined in the data mapping files.

In this two-part series, we walk through the hands-on process of building four distinct Tableau histograms. Once the technical work is done, we leverage AI to tell the story, using Google Gemini to help interpret the distributions and uncover the "why" behind the data.

Building an AI-First Data Architecture (starting with Tableau)

The biggest hurdle for AI in the enterprise isn't the algorithms—it’s the meaning.

If you ask an AI, "What happened to sales?" it can look at the data. But if you ask, "Why are our 3-month rolling averages dipping?" the AI needs more than just a table; it needs to understand the Semantics of that calculation.

I’m using a four-tier metadata framework to bridge the gap between "Raw Code" and "AI Intelligence." This isn't just about documenting for humans; it’s about creating a machine-readable map of Business Intent.

Create a Tableau KPI Dashboard with Current Month performance, calculating a rolling 3-Month Average and a comparison to the previous month.

A comprehensive, 15-chapter roadmap designed to build a strong technical base in data visualization. Free to use, free to learn.

If you’re relying on Python to clean, transform, and aggregate your messy source data, you need to be absolutely certain your scripts are robust, fast, and, most importantly, error-free.

Download my foundational coding guide: Python Coding One Year Later—for free! The focus is on fundamentals and debugging examples.

This challenge isn't about complex formulas; it's about the high human cost of poor data ethics and bias.

Ditch boring tutorials—challenge yourself with a fun game, the ultimate way to master data skills! Dive into live workbooks for every question to test your solution, or follow the "Clues" for a guided path to conquer the puzzle.

https://www.tableauatwork.com/vizwhizleaderboard

In this inquiry-oriented challenge, follow your own path to explore the data clues and solve the case.

Ready to build the dashboards that get noticed? This high-level video explores the Fun Fab Five -  5 essential concepts that unlock advanced visualization: Filters, Aggregation, Level of Detail (LOD) calculations, Parameters, and Actions/Tooltips.

In this video, we'll explore the fundamental roles your data fields play, from Dimensions and Measures to the famous Blue vs. Green distinction of Continuous and Discrete fields.