Senior Product Designer | LLM Interfaces · Behavioral Design · Analog and Systems Thinking | 13 yrs on the craft. Turning non-deterministic systems into experiences people can rely on.
For thirteen years I've designed interfaces where people had to rely on systems they couldn't check — black-box analytics, probabilistic data, now LLMs. I make those systems legible enough to trust, and correctable when they're wrong.
Ask Markerr did not ship to real users. It reached POC/MVP stage and was demonstrated to partners and stakeholders as part of Markerr's AI roadmap. That distinction matters, especially in AI product design, where the gap between demoed and validated can be large.
I led the UX for an AI system that helped government auditors write complex legal instructions. By reshaping messy inputs and giving auditors full control to review and correct the AI's output, we cut writing time by 63%: months to hours. The system shipped, was adopted across teams, and won Best technology case in the public sector.
B2B platform architecture and multi-surface workflow design for HR managers and brokers.
Fintech workflow separation and analytical surface strategy. Replacing spreadsheet-as-product behavior.
Empathy and the drive to earn genuine trust are not values I added to my practice. They are the foundation it is built on. Everything else — the input architecture, the behavioral translation, the code proximity — comes after that.
When little is on the line, people over-trust fluent answers, so you add friction. When a lot is on the line, they're already skeptical, so you give them evidence they can check.
Trust is a design problem, not only a model problem. I reshape the inputs a model sees and give people the right to correct what it produces.
Not the happy path. I design for non-determinism, graceful failure, and correction flows before I design for success.
I build high-fidelity interactive prototypes with AI-Assisted Coding. I stay through implementation instead of through static frames over the wall.
Users don't live in your product; they live in everyone else's. I reuse what they already know so the novelty budget is spent only where the AI genuinely breaks the paradigm.
My approach relies on a proprietary framework grounded in problem-solution co-evolution: Parallel Trust Design.
By balancing Sensemaking (immersion, debate, and philosophical framing) in tandem with an executable Trust Design Toolkit, I use visual pattern-finding to bridge the gap between abstract strategy and concrete product actions. I don't wait for requirements to stabilize; I design to establish Calibrated Trust.
// Interact with filters or focus a canvas node to inspect its methodology parameters.