Conversational AI Advisor
Designing a conversational AI triage and quality framework for an online technology provider — from quantitative log analysis through to the behavioural specification used as the engineering build brief.
View Case Study →Steeve Royce — UX designer
Most recently, that's meant going further than most UX work does — into how AI refuses, defers, and handles uncertainty, not just how it looks on screen.
Work that starts with people and follows the problem wherever it goes — including places most designers don't.
Designing a conversational AI triage and quality framework for an online technology provider — from quantitative log analysis through to the behavioural specification used as the engineering build brief.
View Case Study →
A comprehensive thesis project developing an AI-augmented mobile app with smart sensor integration to help home gardeners achieve consistent yields through data-driven insights.
View Case Study →How should an interface present probabilistic, model-generated analysis without overstating its confidence — when the user might put money behind it?
Can LLM-assisted research analysis be made hallucination-resistant enough to base design decisions on? Citation-forcing plus manual verification was the answer.
Can delight and honesty coexist — attention-grabbing motion that never misleads, on a site about designing trustworthy AI?
Most AI products are designed to project confidence. But confidence without calibration is a UX failure — and eventually a trust failure. Designing for uncertainty isn't a safety nicety, it's what makes AI worth depending on.
Read →
There's a gap between the interface you designed and the interface your users experience. Understanding foveal vision, change blindness, and inattentional blindness changes how you think about hierarchy and placement.
Read →I translate research into design decisions people actually feel — from accessibility reviews to the small behavioural details that make an interface easy to use. Lately, that same practice has gone one layer deeper: evaluation and adversarial testing on the prompts, context, and guardrails that shape those decisions.
MA UX Design (Distinction)
Get in Touch →Always interested in the hard problems where design and AI meet. If you're working on one — reach out.