Behind the work / Norbert Barna
A story in
motion.
People. Products. Perspective.
It starts with
a question.
Seventeen years in, I still work from one question. I learned to ask it inside enterprise security software, where I spent six years: what does a person need to see to act with confidence when a wrong move costs something real? Balabit became Balasys and then One Identity, and through all three names my job stayed the same. I made complex systems legible to the people who were responsible for them.
Everything since has stood on that ground. When trust is the whole product, clarity is what keeps people safe, so I learned early to take a confusing screen as seriously as a broken one.
The question travelled well. It has served me in banking, in enterprise learning and now in capital markets.
Continue to perspective
The work between
the milestones.
02 / Perspective
A wider view.
At Raiffeisen Bank International the question met scale. I built and led a seven person in house design team and set the direction that took the banking app from 1.8 stars to 4.8 and past 300 million euros in transaction volume. The design kit and UX guidelines we established went on to serve teams in 16 countries: one standard instead of sixteen interpretations of it.
At Instructure the question met AI. I introduced AI assisted design to Catalog, the first time the company had worked that way, and in a single year the team shipped 42 features to production. Leading feature development on a product that several million people used every day and that brought in more than 52 million dollars in annual revenue, I set how its AI features were designed and built and held them to usability, accessibility and ethical standards, because an AI feature people cannot review is a liability with a nice interface.
At BlackRock the question meets the market itself. I am a Vice President there and lead a product that manages 15 trillion dollars across its full portfolio. My team owns its AI development and the design of its features, and I run delivery end to end for fixed income trading so global desks can source, analyse and execute liquidity in one place.
The industries changed. The person on the other side of the screen did not. Someone has to make a good decision quickly, with something at stake, and that is who I build for.
Explore the work
What seventeen
years taught me.
I start where the work actually happens. At OnRobot the design that mattered came from the shop floor, from watching operators rig a task and seeing where they guessed or checked with a colleague. At Raiffeisen it came from interviews and analytics before anyone drew a screen. I want to see the problem with my own eyes before I form an opinion about it.
I decide what is worth building before time, money and trust are spent. At Raiffeisen that was a rebuild or patch workshop with stakeholders before we committed to a full redesign. At Instructure we set the guardrails on day one. A person stays in the loop, AI output is labelled and can be edited, and there is always a way back.
I build teams so the work keeps shipping after I step back. At Raiffeisen I stood up the team in five weeks, then gave it OKRs, two-week sprints and reviews where Product, Engineering, Compliance and Support sat at the same table. At Instructure we reviewed flows with Legal, Compliance and Data Science before build and shipped behind flags, and the team still put 42 features into production in a year.
Continue to what's ahead04 / What's ahead
The next
chapter.
I want more of the work I have described here: AI that a person can check before acting on it, products that hold up under regulation, and teams that keep shipping after the launch. Kineticare, a Hungarian digital health platform I designed and built end to end, is what that looks like when I start from zero.
If you are working on something like that,
I would like to hear about it.