Smartlinx replaced legacy QA with AI agents

Company Overview

• Company: Smartlinx
• Industry: Healthcare SaaS / Workforce management
• Business Type: Workforce management software for healthcare providers
• Size: 200–500 employees
• Headquarters: USA

Key Use Cases

• Automated regression testing
• Complex UI workflow validation
• AI-assisted test generation (via Confluence Docs)
• Shift-left developer PR testing

The Challenge: A decade of legacy tests broken by migration

For over a decade, Smartlinx built a massive testing portfolio that eventually swelled to 4,600 UI test cases executed bi-weekly. However, a migration to Angular severely disrupted their existing automation framework, forcing the company to rely heavily on a team of seven offshore manual regression engineers to maintain product quality.

Because modern UI technologies dynamically fetch data through APIs, the Smartlinx team was working to build a resilient backend API framework. However, their product still required heavy UI-based validation for workflows that cannot simply be tested via API. Running these manual regression tests was not only costly and time-consuming, but also susceptible to human error. Smartlinx needed a modern, AI-driven automation solution to replace their legacy framework without having to manually code thousands of new scripts from scratch.

The Solution: AI agents for complex UI and rate limits

Smartlinx partnered with QA.tech to rapidly build out a new, intelligent UI automation suite. Using QA.tech’s agentic AI, they are automating the complex functional layers of their non-prod and staging environments.

Handling complex system states:

The Smartlinx application is deeply complex, requiring different login credentials for specific user rights and careful navigation through various organizational levels (like navigating from Nursing to Dietary departments). QA.tech helps manage these states by reusing specific cached login sessions across tests to avoid backend rate-limiting, ensuring bulk test suites run smoothly.

AI-Assisted test creation from documentation:

‍To speed up test creation, the team leverages QA.tech’s AI chat. By importing summarized feature documentation directly from their secure Confluence pages, Smartlinx can prompt the AI to autonomously generate relevant test steps for new features.

The Results: Repeatable automation and developer enablement

With QA.tech, Smartlinx has already built out over 480 test cases, including a heavy "PPV" regression suite containing over 230 test cases.

By successfully shifting these UI testing burdens to QA.tech's AI, Smartlinx is significantly reducing its dependency on manual regression testing.

"Manual is always prone to errors, right? Versus what it is, it gives you that guarantee that it's a repeatable process, that there is no human bias to change the steps."
‍–  Rajeev Patel, Smartlinx

Looking ahead, Smartlinx is preparing to implement QA.tech directly into their GitHub pipelines. By running AI-generated test cases against pull requests (PRs) as soon as developers submit them, Smartlinx plans to catch bugs before they even reach staging – ultimately enabling developers to build and test faster while saving significant QA overhead.

‍