AI-Driven Testing In Production
DevOps as a culture, movement and philosophy is leading to an increase in the practice of shifting testing to the right — towards production. Many organizations now use continuous integration and delivery pipelines to make decisions about production readiness and, once the software is released, leverage real-time monitoring for detecting and debugging issues. Testing in Production (TiP) has historically been the subject of great scrutiny due to its frequent association with insufficient pre-production testing. However, when applied appropriately, TiP can be a highly effective means of validation and verification. Unlike testing in a lab or staging environment, TiP provides feedback on system behavior using real user scenarios, data and configurations. Tariq King believes that the next-generation of test automation involves combining TiP with AI-driven testing techniques. In other words, the machines of the future will learn how to test by training on information gathered from real users acting in production environments. Similarly, test scenarios will be executed in production environments to provide best effort simulations. Join Tariq as he explains different approaches to TiP with AI, its key benefits and challenges, and how he envisions these technologies moving us towards a future where systems and services test themselves.