I design AI products that work in complex, high-stakes environments.
At Eleos Health, I design AI systems for behavioral healthcare. My work spans documentation, compliance, and revenue, but the core challenge remains the same: ensuring the system is understandable, trustworthy, and usable in practice.
Design Approach
Models can be remarkably capable. The failure point is almost always the interface, how output is presented, verified, and acted on by people with real stakes.
I've spent my career designing the experience of AI, not just screens, but the feedback loops, confidence cues, and review flows that determine whether someone trusts a system enough to use it.
What I find most interesting isn't the model. It's the moment a person decides whether to act on what it says. That's where the real design problem lives, and it's what I keep coming back to.
Live Quality Assist
The central design problem wasn't the UI. It was timing. Compliance review happened months after submission, when nothing could be fixed. I designed LQA to move that check into the moment of writing: nudges over blocks, factor-level reasoning over scores.
Catches documentation issues at the moment of writing, not weeks after submission
Embedded Audio
Invisible when it works, impossible to miss when it doesn't. I designed the recording flow and five distinct error states, each requiring a different response, for a context where a silent failure means finishing a 60-minute session with nothing to show for it.
Became one of the most-used features in the company at launch
Coding Back Office
Providers undercode not from carelessness, but because documentation complexity makes the right code genuinely hard to know. The design challenge: surface AI-suggested codes with enough reasoning to be trusted, and enough transparency to survive an audit.
Surfaces coding gaps hidden in existing billing data
Meridian
A geospatial intelligence prototype for law enforcement communications. Coverage maps, signal analysis, and fleet management — designed to handle complex, high-stakes data without sacrificing speed. Built as a self-contained HTML prototype with no dependencies.
Working prototype — open to inspect, modify, and run in browser
Spent
An AI financial sense-maker that connects spending patterns to the context of your life. Four interaction design patterns, a working prototype, and a transferable point of view on how AI earns trust.
4 AI interaction patterns documented with enterprise parallels
Get in touch
I'm interested in teams tackling meaningful AI challenges in complex domains, and who want a designer who thinks beyond the screen.
carlyraizon@gmail.com