After a few semesters of teaching SQL and python for analytics workshops to total beginners – here’s what I’ve learned. My absolute favorite tool for teaching is Hex because it allows me to: 

• Skip tedious setup steps, like installing packages, which can be unnecessarily frustrating for beginners

• Provide students a worksheet to follow, with instructions, examples, and coding exercises all in one place 

• Give them tools to easily check their work along the way (make a quick chart with a few clicks) 

• Demonstrate to them how to use AI in their workflow while building their analytical skillsets

Here are my top tips for teaching a data skills workshop with Hex: 

Prep and organize class materials

• For each class session or unit, create a Hex project pre-loaded with dataset files, code to load the datasets into dataframes, and an outline of the content and exercises for the class.

  • Add code cells with template code or fill-in-the-blank exercises, using comments to explain what to do next 

• Create an allowed domain for your workspace, then generate and share an invite link for your students to join → saves time on individually inviting everyone

• Create a Collection of projects for your workshop, and share that collection with the workspace → basically functions as a shared folder so you can easily keep track of what’s shared with your students vs. your individual projects

• Make sure students know how to duplicate a project → they’ll need to make their own copy of a project in order to fill in their queries and code 

Show students the ropes

• Start with just the basics to avoid overwhelm: how to run a cell, run the project, and re-run a cell or project to update the cells downstream

• Demonstrate how to find the dataframes in the data browser. Encourage them to check this at each step to make sure that the number of rows and the columns in the dataframes from each cell make sense for the context of the project 

Make class as interactive as possible 

• Break up rounds of lecturing with as many interactive exercises – my goal is to get students writing as much code during the workshop as possible

• If a student runs into an error message, ask them to read it out loud, and walk them through the troubleshooting process. This has always been an essential skill and remains so even alongside AI – plenty of AI-generated code returns errors, too :) 

Share and collaborate

• For group projects, students can easily make one another collaborators and avoid version-control issues by working in the same project

• Ask students to share their work! When working on a data visualization project, ask students to publish their app and share the app view with classmates. This allows me to share students’ work to the class for discussion and feedback (they can also comment on each other’s work!)

Meet students where they’re at with accessibility and AI tools 

• For maximum accessibility, write out as many instructions and examples as you can, in Markdown cells within your Hex templates

  • I used to convey a lot of information via slides while teaching, and I used to get a lot of requests from students to go back to previous slides. In one class I noticed many students would take a photo of a slide and then use Google Translate to translate the text into another language – and I realized this must be slowing them down! The following semester I started writing out all of my slide materials and examples in Markdown cells, and students now seem to have a much easier time following along. 

• You can turn Hex’s AI features on and off at the workspace level → I leave them off at first while students learn the basics, and then I switch it on and show them how to utilize it as part of their workflow. (I miss when this was called “Hex Magic”! It reminded me of the excellent movie General Magic.)

Make it fun

• Bring in emojis, fun datasets, and gifs 😎 In my classes I’ve used Netflix datasets, the dataset of every line of dialogue from Seinfeld, results data from the 2024 Olympics, and some fictional datasets representing real-world case studies. (Tip: Claude/ChatGPT are very helpful for generating realistic-but-fictional datasets in specific formats.)

• Create data apps, not just dashboards, by showing students how to include Input cells and design for their audience to explore

• Teach data formatting and presentation basics alongside the analytical concepts – which are just as important, but often get disregarded or taught in fully separate courses. I think it’s most valuable to teach these skillsets side by side, to give students a real taste of how they’ll use them in the real world

Recommend resources for students to continue learning

A few resources I like to recommend to students:

• learn.hex.tech – easy-to-follow guides specific to Hex 

• Miss Excel – for adding some Excel/data learning to your social media feed 

• Knight Lab SQL Murder Mystery – the most fun online tutorial for SQL

• PolicyViz – data visualization tips and strategies