First Gen-AI Support Assistant in AutoCAD

Lead Interaction Designer, Conversation Designer

DELIVERABLES

Competitive Analysis, Brainstorming Workshops, Early Concepts, Wireframes, Prototypes, User stories

TEAM

AutoCAD Product Team (Product Management, Engineering Design), Customer Success Stakeholders, Technical Support Agents, Autodesk Assistant Team (Product Management, Engineering, Quality Assurance, Data Science, Analytics, Content Strategy, Program Management,Visual Design, Research, UX Designer)

DURATION

5 months

Context

Autodesk’s flagship product, AutoCAD, is a 2D and 3D computer-aided design (CAD) software that is trusted and used by millions across the world. When it comes to troubleshooting an issue in the product:

AutoCAD users attempt self-service by leaving the product context to access resources like Google, AutoCAD Support site, Product Help pages, and the Forum Community for troubleshooting issues with their product.

They submit hundreds of technical product support cases every month but have to wait to hear back

We know from industry research that customers expect fast answers so they can get back to work quickly.

Problem Statement

How might we help AutoCAD customers find answers to their product questions and contact support all in one convenient location so they can get back to work quickly.

Approach

The plan was to leverage the Universal Help module, an extensible, modular help platform, that uses entity-based search to present relevant articles for a user’s issue.

Launch Beta: Update the module to be more chat-like and present top links for their product troubleshooting questions and connect to live chat agents. The Beta version of the bot was added to product help pages that product could users could access from within product.
Learn Fast: Conduct user testing and monitor interaction patterns to tweak and improve the bot’s responses and user experience over time as users adopted it within AutoCAD and other related products.

**** DELETE ****

Update to Generative AI: The Beta version was successful in that it helped us learn and tweak the experience and accuracy of its search results, but the engagement was low as the bot was hard to find. So I partnered with the lead Researcher and Product Manager to craft a vision that convinced AutoCAD team to embed the chatbot with the product canvas thus making it available at their fingertips. In early 2024 we later rebranded the bot as the Autodesk Assistant and The last iteration of this evolution I worked on involved evolving it to new RAG technologies to develop the first Generative Q&A chatbot that speaks “AutoCAD”.

BRAND MISMATCH: Universal Help was visually aligned with the Autodesk web branding guidelines and did not fit the AutoCAD design system that millions of customers are familiar with.

Beta Version

Access via Product Documentation

There was concern that embedding it directly into the canvas would take up valuable real estate so we started with a floating icon in the product documentation pages where some users start their help journey.

Indirect Access

Concern that a floating icon in the product canvas would take up valuable real estate so access to the bot was only via product help pages.

Indirect Access

Concern that a floating icon in the product canvas would take up valuable real estate so access to the bot was only via product help pages.

Indirect Access

Concern that a floating icon in the product canvas would take up valuable real estate so access to the bot was only via product help pages.

Indirect Access

Concern that a floating icon in the product canvas would take up valuable real estate so access to the bot was only via product help pages.

Discovery

I facilitated workshops with stakeholders from sales and e-store organizations to align on top customer intents. I also helped label 200 sample inputs for the data science team to train the AI routing models.