Image

HAX Workbook

Drive team alignment and plan for implementing human-AI interaction best practices early.

What is the HAX Workbook?

The HAX Workbook is a tool for structuring early conversations across the multiple roles needed to implement the Guidelines for Human-AI Interaction.

How do I use the HAX Workbook?

Use the Workbook early in the requirements definition stage of planning a new product or when redesigning an existing one. Because implementing human-AI interaction best practices impacts the entire system, it’s best to bring UX, AI, project management, and engineering together when using the Workbook.

The Workbook is available as a simple and flexible Excel spreadsheet. Download the Workbook to guide your team through five steps:

  1. Selecting which guidelines are relevant to your planned system or feature.
  2. Estimating the UI, AI, data, and engineering requirements needed for implementing high-impact Guidelines. The HAX design patterns can help with this step.
  3. Starting with high impact guidelines, outline UI, AI, data, and engineering requirements for implementing each guideline and estimate the commitment using T-shirt sizing.
  4. Prioritizing Guidelines to implement, considering trade-offs between user impact and cost.
  5. Tracking your progress.

Where can I get the HAX Workbook?

Why the HAX Workbook?

While engaging with product teams, we observed that teams were most successful when they brought all disciplines to the table and planned for implementing the Guidelines for Human-AI Interaction early in the development process.

Teams told us that “a lot of the Guidelines are based upon the engineering more than the design,” and because implementing the Guidelines impacts the entire system, including the data and model, “if the spec doesn’t have that built into it, it’s gonna be too rigid to respond.” We created the Workbook to empower teams to collaborate and plan across disciplines to implement human-AI interaction best practices.

What teams say about the HAX Workbook

The Workbook was the only session where all the disciplines got together to discuss the machine learning model. It gave us a single framework for vocabulary to discuss these problems. It was one of the best uses of our time.

Senior data scientist

It helped us all—data science, product, and design—collaborate in a way that abstracted us from all the technicality of implementing machine learning. We can talk about these very technical things but it comes down to what it means for the user. We’ve integrated the Workbook into our AI process.”

Principal data science manager