Resume
Having a long day?Pink noise, even across every octave with a slow 10 Hz swell, takes the edge off the room. Attention lasts longer when less is competing for it.
email copied
  • Product Design
  • Human-AI Interaction
  • Human Factors

I design trustworthy interactions for safety-critical AI systems.

Seven+ years across design, now in safety-critical AI. Looking for product design roles in AI, robotics, healthtech and mental health.

In depth

01 / 03

Multi-modal sensory feedback · LLM uncertaintyAI sounds sure even when it's wrong. A research-led signal marks when to look twice.

One answer, two things it might be, and the signal that reaches the human by another channel A node labelled AI answer branches at a marked point into two outcomes, true and invented, drawn at the same size and weight and joined by the label identical on screen. Both outcomes then converge again and arrive at a single node: human judgment. A separate solid line leaves the branch point, travels the long way round beneath the whole flow, and enters that same node from below. It is labelled signal. The text channel cannot distinguish the two outcomes; the second channel is what does. AI answer True Invented Identical on screen Human judgment Signal
A phone chat in its warning state: the answer approves 800 mg of ibuprofen, and the row beneath it reads Reliability, Low. The reason has opened under the answer.
Role
Sole researcher and designer
Scope
Self-directed research and system design
Practice
Human-AI Interaction · Human Factors
02 / 03

Human-first NLP · People MatchingBuilding enough safety for someone to type the thing they actually want.

Intent overlapping into a system, through to a meetup Four equally sized circles. Intent and match overlap - a dot sits at the centre of the hatched intersection, with a line rising from it to a caption above the circles reading validation gap, written out in full rather than fitted inside the small overlap itself. Below both circles, a dashed bracket spans the full width of the venn - intent's left edge to match's right edge - labelled NLP engine, showing that mechanism is what produces the whole overlap, not just the gap. From match, a line continues right, labelled progressive disclosure on two lines above it, to the anonymous reveal circle - the one node drawn filled and in ink, because it is the mechanic the study turns on - then on to meetup with no further label, matching the source's own last, unnamed connector. No circle is drawn larger than another - the overlap comes from how close intent and match sit, not from their size. Intent Match Validation gap NLP engine Progressive disclosure Anonymous Reveal Meetup
Tribe product interface
Role
Lead designer — solo
Scope
MVP · design to handoff
Practice
Product Design · Human-AI Interaction
03 / 03

Career Anxiety Research · Dashboard DesignGiving someone a clear next move when the future feels automated away.

The data was the threat Two paths leave one point as more data is shown: the expected one rises towards an answer, the one that happened falls into threat. More data Answer Threat
Mindure dashboard
Role
Behavioral Researcher & Designer
Scope
Deep research · model prototype
Practice
R&D Human Factors · HAI

In brief

Three design variants resolving the same manufacturing constraints differently — logo clearance, hole radius, corner tolerance. Proof that design decisions have physical consequences.

What
Bracelet design for camera straps
Status
Shipped and manufactured
Year
2024
Context
Zooom Productions

No prior users. Wireframed entry flow, dashboard architecture, and deep business analytics from first principles — cognitive load, trust, orientation under uncertainty.

What
Low-fi wireframes, from business registration to a full dashboard with deep analytics
Year
2025
Context
Client work

Making legal frameworks and public grants approachable for the families who need them the most — but get lost before they ever reach the application.

What
Built end to end in WordPress
Status
Fully shipped
Year
2026
Context
For Our Families Foundation · client work, freelance
Visit live site ↗
Grant eligibility interface

Who I am

It's Orsolya.

— means little bear.

I'm a product designer in human-AI interaction and human factors. I work on what happens when a system doesn't behave the way someone expects.

I came to software from live television broadcasting—a zero-error-margin, high-stress setting with real-time graphics—and then graphic design. Later, three years in rural and tribal communities across Australia and Asia showed me how profoundly environment shapes behavior. Together, those years are why I see cause and effect in behavioral ecosystems and structural problems long before I see the screen. I build what I design, with AI tooling.

Working with me, you hear what's wrong early rather than late—and what we can ship in the very same sentence. I never mind diving back to first principles.

What shapes the way I design

Human Factors & Usability Engineering Deepened my grasp of cognitive and physical limits, and how to analyse risk in a design. Arizona State University · Specialization Certificate · 2026 Systems Engineering Gave me a framework for cause and effect — how to hold a complex system in view and steer it. MathWorks · Certificate · 2026 Philosophy of Cognitive Science Sharpened why attention and timing decide what information lands and what never arrives. University of Edinburgh · Certificate · 2025 Foundations of Human-Centered AI Taught me to design trust into AI systems, rather than expect accuracy to earn it. Northeastern University · Certificate · 2025