Flex Assistant

Business context

Enterprise workforce management involved fragmented tools, manual workflows, and low trust in automated recommendations. Leadership needed a single AI-assisted experience that could scale across regions without compromising compliance.

Challenge

The team faced strict regulatory constraints, sceptical end users, and a compressed delivery timeline. Early prototypes felt generic — users did not trust outputs they could not verify or correct.

My role

• — Created product vision for AI-assisted workforce workflows
• — Aligned stakeholders across product, engineering, and compliance
• — Led discovery and shaped UX strategy for trust-first AI patterns
• — Coached designers on conversational UX and human-in-the-loop design

Approach

We ran structured discovery with operations teams, mapped high-friction workflows, and prototyped assistant interactions with explicit confidence signals. Iterative testing focused on trust, not speed.

Outcome

The assistant reached 80% adoption among enterprise users within six months. Support tickets for routine workforce tasks dropped measurably. The design team established reusable patterns for AI transparency across the product suite.

Reflection

Trust is a design problem before it is a model problem. Teams that invest in explainability and user control ship AI products that last.