I design AI systems for high-stakes contexts, where the interface earns trust, or people get hurt.
My work sits at the edge of what AI product design usually covers: non-visual interaction, safety-critical feedback loops, and the hard problem of making machine output legible to humans under pressure. The interface is the contract between human judgment and machine output, that's the principle behind every product decision I make.
I work end to end: problem framing, user research, interaction architecture, and technical prototyping. I'm strongest where design meets real constraint, on-device inference limits, sub-500ms latency budgets, multimodal redundancy for users who can't rely on a single sensory channel.
Before any of that, I spent two and a half years shipping a paid education platform to 50,000+ adult learners across Southern Africa with no formal research training. That gap is what took me to Iowa State for an MS in Human-Computer Interaction, and why I care more about validated evidence than polished mockups.
My background in journalism and media production isn't a detour, it's a foundation. Knowing how people process information under pressure, and how to earn attention in high-noise environments, transfers directly into voice interaction, alert hierarchy, and cognitive load in safety-critical systems.