Building an AI workflow platform for clinical operations by Pablo CorsoBuilding an AI workflow platform for clinical operations by Pablo Corso

Building an AI workflow platform for clinical operations

Pablo Corso

Pablo Corso

Overview

Polytrial is a SaaS platform that enables clinical teams to build, validate, and share AI workflows aligned with regulatory frameworks (FDA, EMA, ICH). The challenge was to create a system that makes AI usable, reliable, and compliant in high-stakes environments.

Design System & UI Foundation

Built a scalable design system (100+ components) to support a complex product surface, ensuring consistency, accessibility, and faster development. Established reusable patterns for structured workflows and data-heavy interfaces.

AI Workflows & Product Experience

Developed core features for building and managing AI workflows, including prompt structuring, validation flows, and user-friendly interfaces for non-technical users. Focused on making complex, regulated processes intuitive and usable.

Real-Time Collaboration

Implemented real-time collaboration features using Liveblocks, enabling teams to build and refine workflows together with synchronized updates and shared context.

Validation & Regulatory Frameworks

Contributed to features that allow users to validate AI outputs against regulatory frameworks (e.g., FDA, GCP), helping ensure reliability and compliance. Designed interfaces that clearly communicate validation logic and results.

Performance & Accessibility

Ensured high performance and accessibility across critical flows, including keyboard navigation, focus management, and screen reader support—key for enterprise and regulated environments.

Tech Stack

React · TypeScript · Next.js · Tailwind · Liveblocks · Storybook · Supabase
Like this project

Posted Apr 4, 2026

Built scalable frontend architecture, real-time features, and a 100+ component design system using React and TypeScript, with strong UX focus.