There are certain aspects of Zebrunner Testing Platform that require improvements. The initial integration with existing test frameworks, particularly for more customized TestNG configurations, requires non-trivial setup work, and the documentation does not cover all the edge cases we encountered. Zebrunner Testing Platform's own test runner, the Carina framework, is well supported by teams using other frameworks but delivers a somewhat degraded experience in terms of deep feature integration. The platform's resource provisioning for parallel test execution can occasionally result in queuing during peak periods. When running large test suites, though this has improved over the past year, the pricing model based on parallel execution slots requires careful sizing to balance cost against test execution time. Regarding Zebrunner Testing Platform's AI capabilities, I think that governance and security could be better. After all, it is a SaaS platform, and it could be better if they provided on-premises deployment support. On a SaaS platform, you cannot fully trust it overall, but it is acceptable as long as the organization is comfortable with it.
