← All work
Design systems & deliveryAUTODESK / CASE STUDY

Revit Weave WPF Toolkit

ABI - BID - Revit, Experience Design

Building a Unified Design System for a 20-Year-Old Powerhouse

My Role: 

UX Designer & Researcher on the core team. I led the foundational research, explored the component difference in different frameworks, created component variants, built the initial component library, and created the documentation and processes to enable pilot teams to adopt the new system.

Key Contributions: 

Design System Strategy, Technical Discovery, UX Research, Component Library Design, Developer Enablement, Documentation.

The Challenge

Autodesk Revit, used by millions, suffered from a fragmented UI built on four different technologies (MFC, WinForms, WPF, Qt). This legacy created inconsistent user experiences, a maintenance nightmare, and blocked compliance with Autodesk's company-wide Weave Design System.

First Thing First - What is Weave Design System?

Autodesk Weave is the unified design system that provides a comprehensive set of design, content, and interaction patterns for all Autodesk products. It aims to enhance consistency, accessibility, and brand alignment across various applications, enabling designers and developers to create cohesive user experiences efficiently.

01

The Problem: A Fragmented UI Ecosystem

An Inheritance of Inconsistency

Over two decades, Revit's UI had become a patchwork of technologies, each with its own look, feel, and code patterns. This fragmentation had real consequences:

  • For Users: Inconsistent and sometimes jarring experiences when moving between different parts of the software. No dark theme support across the board.
  • For Developers: Immense overhead. Teams wasted time "reinventing the wheel" on UI, fighting with outdated tech (MFC), and maintaining three different journaling systems for regression testing.
  • For Autodesk: An impossible barrier to delivering a cohesive, modern brand experience across its entire product suite, as mandated by the Weave initiative.

How much work?

Our analysis found over 600 individual dialog boxes built on legacy technology that needed to be modernized.

02

The Strategic Decision: Why a WPF Toolkit?

Evaluating the Path Forward

We evaluated multiple strategic options, from migrating everything to web-based React to creating a completely abstracted UI framework. Our chosen solution had to balance technical feasibility, developer efficiency, and the aggressive FY26 deadline.

Why WPF?

  • Strategic Alignment:
    WPF was already Revit's strategic .NET UI technology, recommended to third-party developers.
  • Capability:
    WPF's powerful styling system could faithfully implement the entire suite of Weave design tokens (colors, typography, spacing, shadows).
  • Pragmatism:
    It allowed us to build a bridge. We could style new WPF dialogs immediately while creating a path to eventually migrate older MFC/WinForms dialogs.

Our proposal was to build an official, company-sanctioned Weave WPF Toolkit—a library of pre-styled, reusable components that any Revit team could use to build compliant UI quickly.

Decision Matrix

Technical Feasibility
Development Efficiency
Low
High
High
Low
Migrate to Qt
Web-based React
Abstracted UI Framework
Build WPF Toolkit

The matrix above visualizes our decision-making process, showing how the WPF Toolkit solution offers the best balance of technical feasibility and development efficiency.

03

The Solution: Pillars of the New Architecture

A Two-Part Solution for Scale and Consistency

We designed a dual-pronged approach to tackle both new and existing UI:

1. The Weave WPF Toolkit (Autodesk.Weave.Wpf)

A NuGet package providing ready-to-use, Weave-styled versions of all standard WPF controls (Buttons, Textboxes, ComboBoxes, etc.). This was the fastest path for new features and dialogs.

2. The GeneratedUI Framework

A more ambitious, long-term architecture that separates business logic from UI code. Developers describe their UI using data patterns (e.g., BooleanInput, ChoiceInput) rather than specific controls. A runtime engine then generates the actual Weave-compliant WPF UI. This is the key to efficiently converting hundreds of legacy MFC dialogs.

This approach ensured immediate wins while building a scalable foundation for the future.

04

Demo of the toolkit

Sample App

We’ve built a sample app which allows not only developers but also designers to test the components in the toolkit and check the Docs directly from the sample app.

Converting a dialog with the toolkit

Below is an example of how the toolkit helps converting a MFC dialog to Weave WPF style.

05

My Role: Leading the Pilot & Proving the Concept

From Theory to Practice: Enabling the First Teams

My focus was on helping the dev team turning the strategic vision into a practical toolkit that development teams could actually use. Key contributions included:

Component Design & Specification

Worked directly with engineers to define the API and behavior of foundational WPF controls. For example, I specified the correct margin (8px) for dropdown icons when a token discrepancy was found, ensuring pixel-perfect alignment with Figma.

Quality Assurance & Design Review

Served as the final design authority, reviewing and approving pull requests to ensure visual fidelity. I verified shadow effects, background colors for DataGrid headers, and TextBox styling before they were merged into the main library.

Developer Experience (DX)

Championed developer adoption by creating comprehensive documentation and processes. I authored the "Step-by-Step Guide to Requesting Weave WPF Design Tokens" to streamline intake and built a "test app" catalog for self-service exploration.

Pilot Program Management

Onboarded and supported the first pilot teams, gathering critical feedback that shaped the toolkit's evolution. I acted as the bridge between developer questions and the design system's truth.

Bridge Building

Facilitated communication between the Revit team and the central Autodesk Weave team to ensure our WPF implementation was recognized and supported.

06

Validation & Impact: The Developer's Verdict

Measuring Success: A Report Card from the Teams

After the initial rollout, I designed and conducted a survey of the development teams using the toolkit to measure its effectiveness and identify areas for improvement. The results were overwhelmingly positive.

Quantitative Impact

  • 4.25/5 Average Satisfaction Score
  • 50% Average Time Saved on UI development
  • 5+ Product Teams successfully onboarded (Revit, Inventor)
  • 90%+ Time Saved reported by the highest-performing team

Qualitative Feedback

Teams reported high satisfaction with the consistency and quality of the components. The Twinmotion Integration team (Spectrum) and the Manage Links team (Marvel) reported being "Very Satisfied," noting the toolkit saved them 1-2 weeks of development time.

Actionable Insights

The survey also uncovered key areas for iteration, such as the need for better DataGrid examples and improved MVVM support, directly informing our priorities for the next development cycle.、

CONTINUE EXPLORING

Fusion Makersite Plug-in↗