Oryn is a personalized AR-HUD paired with a biocompatible neural sensor. It tracks four mental states 
cognitive clarity, knowledge consolidation, incentive health, and social & stress authenticity.
The goal is to make the unconscious conscious. Enough visibility into your own neural patterns to actively reshape your behavior, instead of being passively governed by it.
I led UX and systems design on a small hackathon team. My focus was the interface architecture: the interaction model and the spatial logic of the brain users would navigate. That meant holding multiple users at once. Someone managing OCD has different needs than an office worker in burnout or a student with cognitive fog. The interface had to work for all of them without collapsing into a generic wellness aesthetic.
My teammates led the visual rendering (color system, brain model, HUD aesthetic), and I helped develop the final pass. Across 83 hours, the collaboration moved fast: UX decisions set the constraints, and visual rendering pressure-tested them.
The FigBuild 2026 hackathon prompt asked for a tool that tracks an aspect of human sensory experience and lets the user influence it, while being speculative, future-forward, and designed for capabilities that don’t yet exist.
We translated this to the most under-measured sensory experience a person has: their own brain. We focused on neurotransmitters and dendritic synaptic connections, the messengers and receiving structures behind neuroplasticity, which shapes how we learn, recover, and behave. These metrics are invisible to us, yet recovery from trauma, addiction, and ingrained patterns all lives in this layer. We wanted to give it an interface.
How might we make awareness possible without making obsession easier?From our earliest discussions, the interface had to make awareness possible without making obsession easier, especially for users whose mental-health histories make over-monitoring its own kind of risk.
The viewport is a field of view, and attention is already engaged with the physical world, so every element competes with reality. Information has to register without being read, and stay simple enough to process mid-conversation or mid-crisis.
A numerical readout on the HUD was the obvious first option, and the one we rejected first. A live number becomes something to optimize against, which is the opposite of clarity. So color fields and saturation carry the four metrics in real time instead. Numerical analysis still exists in the dashboard for historical trends and pattern comparisons, but the live signal stays ambient. Color does the same categorical work without inviting the same compulsion.
Anatomical mapping does cognitive work that simple icons cannot. The user can clearly see where something is happening within their brain, and not just what it is and the spatialization makes the data much more legible without a legend.
Hand gestures and voice commands are the standard AR conventions, and both pull attention away from the world the user is already in. Oryn's interaction model surfaces with the user's attention and recedes without it. The interface is closer to a sense than to a tool or an app.
Prototyped in Figma: the interaction model, the brain model in its ambient and focused states, and the dashboard layer behind the HUD.
People: Users share a simplified, human-readable signal of their mental state with people they choose.
Items: Oryn detects neural fluctuations around everyday objects and triggers contextual environmental responses.
EEG ID: Every Oryn user receives a unique EEG ID, or the personal identifier that gates access to Resonate.
Three guardrails, because extra perception comes with responsibility.
Neural data is encrypted and never shared with external parties without explicit user consent via Resonate EEG ID.
The temptation in any self-quantified product is leaderboards, streaks, and connections. We ruled them out. Brain performance shouldn't be something to compete on, and the moment it becomes one, the product is doing harm.
Engagement is not a metric, and silence is the default. The interface only interrupts the user when the signal is meaningful enough to warrant attention. This is what protects the user from turning the tool into a source of hypervigilance.
Calibration, neural mapping, medical onboarding, and the sensor material itself are real product work we couldn't reach responsibly in a hackathon. A real next pass would treat them as a medical-adjacent moment, not a setup screen. This is where trust is built or lost permanently.
Most of Oryn's strongest scenarios depend on personal historical baselines, none of which exist on day one. That early window is its own design problem. How does Oryn behave when it can show current states but not yet patterns? What does the interface promise, and what does it withhold, while the baseline is still forming? Those are the questions I'd take into the next pass.
I'd love to chat more about my process: