Collective IQ
Overview
Collective IQ is an AI-powered product for Microsoft 365 and Teams, currently in private preview. I joined as the sole designer at the start of the project and have owned the full design process, from early product framing through component-level execution, working directly with engineering and stakeholders across rapid iteration cycles.
Because the project hasn't shipped publicly, detailed case study content is available by request.
What I did
- Led 0-to-1 design on an enterprise-facing AI product embedded in Microsoft Teams, building the end-to-end information architecture from scratch
- Designed complex interaction patterns for an AI-mediated workflow involving multiple user roles, permission states, and asynchronous actions
- Built and extended a component system in Figma using Fluent UI v9, including custom components where the existing library had gaps
- Designed across the full contributor experience: from in-product triggers through confirmation states, receipts, and edge cases
Why this was hard
Enterprise AI UX is a different problem than consumer AI. The people using this product aren't choosing to be there. They're embedded in existing workflows, with limited attention and high expectations around trust and control. Designing for that context meant making AI feel predictable and low-friction while still giving it enough surface area to be genuinely useful.
A lot of the work was figuring out where AI should be invisible and where it needed to earn trust explicitly. That line moves depending on the action, the user's role, and the stakes of what the AI is doing. Getting it wrong in either direction breaks the product's usefulness in ways that are hard to recover from.
Interested in the work?
Let's set up
a walkthrough.
I'm happy to walk through the full case study with anyone in a hiring process or able to sign an NDA. Design decisions, process, artifacts, and outcomes. Reach out and we'll find time.