Role-based dashboard
Separate surfaces for submissions, trial progress, and decisions so each role sees only what they need.
Case study · Healthcare SaaS
Medical device trials, powered by AI
Tryalz is a B2B platform that digitizes how hospitals evaluate new medical products. The goal was simple: replace paper surveys and subjective reporting with a structured, AI-assisted workflow.
- Industry
- Healthcare SaaS
- Product
- Medical device trial management platform
- Core value
- AI bias detection + standardized evaluations
- Delivery
- From prototype to live MVP
- Timeline
- 3 weeks for full trial loop
Challenge
Hospital trials needed reset
Medical product evaluations at hospitals historically relied on paper surveys and verbal reports, leaving room for bias and inconsistency in high-stakes purchasing decisions.
The team needed a platform that could collect feedback objectively, flag biased questions before trials began, and produce a defensible decision trail without adding friction for clinical staff.
Solution
AI-assisted trial workflow
Development started from a working prototype with limited functionality. The first milestone was a stable foundation with role-based access and tailored dashboards.
The next layer introduced the product signature: AI bias detection on evaluation questions, neutral rewrite suggestions, and automated summaries for open-ended feedback. Within 3 weeks, the full loop was live from request submission to final decision.
Product surface
Features built for trust
Bias review before launch
Evaluation questions are scanned for leading or loaded phrasing, with AI suggestions reviewers can accept or refine.
No-login QR evaluations
Clinical staff submit feedback by scanning a QR code or opening a direct link, reducing friction in real trial settings.
Decision snapshot
Per-question yes/no and rating distribution makes trial outcomes transparent and easier to defend in purchase reviews.
AI trial summary
Open and scored responses are synthesized into a neutral summary to reduce subjective interpretation across stakeholders.
Open-ended synthesis
Text responses are grouped into concise themes with context on quality and signal strength of submitted feedback.
Technology
Modern healthcare SaaS stack
The platform was built with React, TypeScript, ShadCN, Storybook, NodeJS, PostgreSQL, oRPC, and TanStack, balancing fast iteration with production stability.
React TypeScript ShadCN Storybook NodeJS PostgreSQL oRPC TanStack
Results
Prototype to live MVP
3
weeks
From prototype to production MVP
3
AI touchpoints
Bias detection, question generation, and response summarization integrated
90
development hours
Production-ready healthcare SaaS with 4 roles, 6 workflows, 3 AI integrations, and Stripe payments
Tryalz moved from a frontend-only demo to a deployed, production-ready MVP within the agreed development scope, ready for real users and real trial decisions.