How Libra Bank reduced test creation time by 35% with UiPath Test Cloud

Bringing AI-powered software testing to banking teams

Before adopting UiPath Test Cloud, Libra Internet Bank relied mostly on manual QA processes to test new developments. Test case creation was done in Xray and was manual and time-consuming. Smoke and regression testing, supported by Jira and Xray, were also largely manual, which made it harder to cover all user scenarios within tight timelines. Some teams were using Selenium, Java, and Appium for automation, but coverage was still limited and had not yet scaled across the bank. Libra was also using UiPath on-prem, with processes and unattended robots supporting some testing activities. As delivery demands grew, the team needed a more efficient and scalable way to connect manual testing, test automation, and AI-driven test design without disrupting existing workflows

To support this effort, Libra Internet Bank implemented UiPath Test Cloud, embedding agentic software testing into its quality engineering approach.

The bank began with a pilot across two teams working on client onboarding projects. The initial goal was to evaluate the effectiveness of AI-generated test cases compared with manually written test cases in Xray. One of the teams already had UiPath-based automation in place, making it possible to assess low-code automation generation and the integration of existing UiPath tests into Test Manager as part of the Test Cloud solution.

With UiPath, Libra implemented a testing approach that supported:

• AI-powered test case generation based on business requirements in Jira

• Integration between Jira, Xray, Planview, and UiPath Test Manager

• Low-code automation development and execution on internal virtual machines

• Integration of existing UiPath automations into a test case-oriented testing model

• Expanded smoke test automation across teams that previously had no test automation

Following strong results, Libra extended AI test case generation across all project teams. In parallel, the bank launched a broader effort to improve overall testing efficiency by expanding smoke test automation to teams that had previously relied only on manual testing.

We consider UiPath Test Cloud a key component in advancing our quality engineering capabilities at Libra Internet Bank, particularly through its AI-driven approach to test design. By simplifying and accelerating the creation of test scenarios, it reduces manual effort and enables our teams to focus on higher-value activities. This contributes to a more scalable and efficient testing framework, aligned with our long-term digital transformation objectives. In addition, UiPath Test Cloud simplifies test execution of automated tests through its user friendly interface and integrates seamlessly with UiPath Studio. “

Sorina Dumitrescu, Director of Projects Delivery Division, Libra Internet Bank

Outcomes

• Faster test case creation: Libra reduced test case creation time by 35% at bank level by using AI-powered test case generation.

• AI adoption at scale: More than 60% of test cases are now AI generated and used in the bank’s testing process.

• Expanded smoke test automation: More than 90% of smoke tests are now automated in teams using UiPath as their software testing automation solution.

• Improved release confidence: While broader compliance and governance impacts are still developing, release confidence has already improved because most smoke test suites are now automated with UiPath Test Cloud.

Building a scalable testing foundation for the future

With UiPath Test Cloud integrated into its software delivery process, Libra Internet Bank has established a stronger foundation for scaling AI-powered software testing across the organization.

And the impact goes beyond automation. UiPath has helped the bank bridge manual and automated testing, allowing manual testers to trigger automated tests directly from UiPath Test Manager in Test Cloud. It has also increased awareness across developers, business analysts, and business teams of how AI can support software development and software testing more broadly.

AI-generated test design is also creating downstream process improvements. Because business requirements in Jira now serve as critical context for AI-generated testing assets, teams are moving toward greater consistency and standardization in how requirements are written and maintained.

Testing is becoming more connected, more scalable, and more aligned with the bank’s long-term digital delivery goals.

From pilot initiative to broader software testing transformation

Libra Internet Bank’s journey shows how financial institutions can use agentic software testing to modernize quality practices without replacing the tools and workflows they already trust.

By starting with a focused pilot and then expanding across teams, the bank improved efficiency in test design, scaled smoke test automation, and created a clearer path between manual testing, automation, and AI. What began as an effort to evaluate AI-powered test case generation has evolved into a broader transformation of the bank’s software testing model.

Looking ahead, Libra plans to continue improving AI test case generation, accelerate test automation rollout with AI, expand automation coverage, introduce performance testing, implement AI-powered self-healing tests, and evaluate agentic AI use cases both in testing and in other software development and business processes.

With UiPath Test Cloud, Libra Internet Bank is not only improving testing efficiency today, but building the foundation for a more scalable, AI-powered software testing strategy for the future.

Learn more about Test Cloud.