Software development is undergoing rapid changes with the increasing use of AI coding assistants and large language models (LLMs) like Cursor and ChatGPT, which enable faster code generation but also necessitate comprehensive testing to ensure reliability. As AI-generated code may not always behave as expected, rigorous testing becomes crucial to verify not only functionality but also the intent behind the code, prompting the use of LLMs for generating test data. These models excel at creating diverse and realistic test cases, enhancing test design by producing a wide array of inputs, including edge cases. An example of this is using LLMs to generate test data for a user onboarding flow, which involves creating varied JSON payloads to test different scenarios. This approach can be integrated into continuous integration and continuous deployment (CI/CD) pipelines, such as CircleCI, to maintain code consistency and quality despite the non-deterministic nature of AI models. By leveraging LLMs effectively in test generation, teams can improve test coverage and efficiency, making it a valuable addition to modern software development workflows.