AI Platform • Monitoring & Observability

Noah

Empowering AI researchers with a robust, high-fidelity monitoring platform for model drift and performance observability.

ultramodern dark mode AI monitoring dashboard, vibrant lime green data visualizations, neural networ
Role

Main Founding Designer

Status

UI UX Intern (Startup Founding Team)

Company

Cortif Technologies

Tools

Figma

The Challenge

As the sole designer for a stealth-mode startup, I was tasked with creating a platform that could visualize complex AI model metrics for technical users while maintaining a premium, "enterprise-grade" aesthetic that builds investor confidence.

Drift Detection

Visualizing the subtle shift in model accuracy over time across millions of data points.

Scalable Systems

Building a flexible design system from scratch that could evolve with the rapid engineering pivots.

Core Dashboard Architecture

wide desktop view of a complex AI dashboard, dark mode, neon green accents, interactive charts for m
screen showing model health status, glowing status indicators, minimalist dark inter
detail of a dark mode UI component library, buttons, dropdowns, and toggles with neon green accents,

The Solution

I designed Noah with a focus on "Data Clarity." By utilizing high-contrast visuals and interactive drill-downs, researchers can identify model anomalies in seconds rather than hours.

The branding was designed to be futuristic yet stable, using a deep monochromatic palette with green accents to signify growth and precision.

scenic view of a tech office at night, designer working on a laptop with the Noah AI dashboard on sc

User Flow

This user flow maps the monitoring experience in Noah, showing how users configure and continuously monitor their AI models and API endpoints. The journey begins with selecting what needs to be monitored, followed by activating monitoring and allowing Noah to continuously evaluate system health and performance. When an issue is detected, the flow introduces decision points for determining whether the issue has been resolved, with users able to remediate, retest, and return to continuous monitoring. Mapping this flow helped clarify the relationship between user actions, automated system behavior, and decision points, while designing an experience that makes complex AI monitoring workflows easier to understand and manage.

Noah user flow diagram showing the monitoring experience for AI models and API endpoints