Back to blog

From a Windows Kiosk to a Tablet Solution: Lessons from an AI System Prototype

Healthcare case study: how a Windows kiosk prototype revealed hardware limitations and led to a shift toward a tablet-first architecture.

Cover image for: From a Windows Kiosk to a Tablet Solution: Lessons from an AI System Prototype
Photo: Slawomir Kocur. Taken on October 9, 2025 during a conference at Ptak Warsaw Expo.
Case studyAIAutomationHealthcareSystems architectureUX

From a Windows Kiosk to a Tablet Solution

What an AI Prototype for Healthcare Taught Us

In IT projects, especially where software, hardware, and end users intersect, the first concept is rarely the final one. One such project was an information system for patients after hospital discharge. We initially designed it as a stationary kiosk and later transformed it into a much simpler tablet-based solution.

This article is not a story about a “failed implementation.” It describes concept validation in real-world conditions, the lessons we learned, and a deliberate architectural decision that reduced project risk and cost.


Project Goal

The goal of the system was to provide patients with:

  • quick access to information after discharge,
  • a simple interface that did not require staff assistance,
  • consistent and up-to-date informational content,
  • room for future expansion with automation and AI elements.

From the beginning, a key assumption was maximum ease of use for both end users and the organization maintaining the system.


Initial Concept: Stationary Kiosk

The initial architecture included:

  • a stationary kiosk based on Windows,
  • a PC enclosed in a dedicated housing,
  • a printer for handing out information,
  • kiosk mode to restrict access to the operating system.

The solution was built as a prototype and presented publicly at an industry conference, which allowed us to test it in conditions close to production.


Problems That Appeared in Practice

Testing and the public demo quickly showed that hardware became the biggest risk factor in the entire system.

Key observations:

  • failures and unpredictability of hardware components,
  • printer issues (paper, drivers, maintenance),
  • long startup times,
  • complex servicing and updates,
  • high maintenance cost compared with actual end-user value.

From both UX and operational perspectives, the heavy kiosk setup created more problems than benefits.


Decision: Change of Direction

Based on these findings, we decided to move away from the stationary kiosk toward a tablet solution.

This was not purely a technology decision, but a business and operations decision.

The tablet approach:

  • eliminates most hardware issues,
  • does not require complex servicing,
  • starts up instantly,
  • is cheaper to maintain,
  • offers better ergonomics for users.

Importantly, the entire system logic, backend, and functional concept remained valid. We changed only the hardware layer and access method.


New Architecture (High-Level)

The target approach is based on:

  • a tablet (iOS / Android),
  • a web app or PWA,
  • a central backend and API,
  • server-side automations,
  • opportunities for deeper AI integration.

This solution is:

  • simpler,
  • more scalable,
  • cheaper to maintain,
  • easier to deploy across many locations.

Key Takeaways

This project confirmed several universal principles:

  • hardware is often the weakest link in IT systems,
  • the simpler the end device, the better the adoption,
  • kiosk != tablet, even if they appear functionally similar,
  • AI and automation matter only when UX is frictionless,
  • fast real-world validation saves both time and budget.

Summary

The public prototype presentation helped us validate assumptions quickly and avoid a costly mistake before full implementation. Instead of pushing a solution that “looked good on paper,” we changed direction based on real observations.

This iterative, fact-based approach, rather than attachment to a specific technology, is what we aim to apply in every project.

If you are considering systems that combine software, automation, or AI with end devices and want to avoid similar pitfalls, let’s talk.


Learn more about our approach to software development and architecture or see all case studies.