Causal Foundry Platform
Design Systems and Data Visualization for AI-Driven Tools
Client
Year
2022-2024
Industry
Healthcare, Tech, AI
My Role
Lead Designer
UX & Design Systems
Related Case Study
The Challenge
Causal Foundry’s AI platform empowers healthcare providers working in underserved communities by offering actionable insights through causal inference models. As the product evolved to include features like behavioral nudges, cohort segmentation, and A/B testing, the existing UI couldn't keep up. The challenge was twofold:
Make complex, AI-powered workflows accessible to non-technical users.
Build a consistent, future-proof design system that could scale with rapid product growth.
I led the UX design of core AI workflows and built a modular design system from the ground up. My focus was on creating a clear interface structure that could scale with future ML capabilities—while ensuring visual consistency, usability, and trust across the platform.
The Foundations
Before diving into design, I mapped the full spectrum of AI touchpoints across the platform—from automated nudges to model-generated cohort suggestions. This gave us a clear view of where human decisions met machine recommendations, and where trust needed to be built.
Together with the ML and product teams, I shaped a product architecture that allowed for guided, step-by-step flows. These helped users understand not just what the system was suggesting, but why—transforming abstract models into concrete, actionable tasks.
The Structure
To support the product’s pace and complexity, I built a design system from the ground up. Starting with component logic, spacing grids, and color tokens, I defined interaction patterns that prioritized clarity, trust, and accessibility. Each element was designed to be modular—so that as new tools emerged, the system could flex without fragmenting.
The design language balanced brand personality with functional restraint—anchoring the interface in a clean, clinical aesthetic without feeling sterile.
The System
Alongside visual and interaction specs, I built documentation and implementation guides that helped engineers plug components into the codebase with confidence. The system covered everything from hover states to dark mode adaptations, and included built-in accessibility guidelines.
We set up a feedback loop between design and engineering via Slack and weekly reviews—reducing ambiguity, speeding up build times, and making collaboration smoother across the board.
The Outcome
The design system became a critical foundation for scaling both product design and development velocity. Engineers could ship faster with fewer handoffs. Designers had a reliable system to build from. And most importantly, healthcare users—many with limited technical literacy—could confidently navigate AI-powered workflows.
The AI interfaces became clearer, more approachable, and ultimately more effective at driving action—without sacrificing the nuance and responsibility required in healthcare settings.
This project taught me the power of pairing rigorous systems thinking with human-centered transparency. By embedding clarity, modularity, and empathy into every layer—from component grids to AI explainers—we helped transform a powerful but complex platform into a tool that real people could trust and use with confidence.
