AI Writes the Code. Who’s in Control?
AI coding agents are writing more of the code that reaches production, including in Java. But when pull requests grow larger and arrive faster, it gets harder to review every change and know who decided it was ready to ship. Join Barcelona JUG and Quality Clouds for a practical session on keeping people in control as agents become part of the development workflow.
Albert Franquesa will show how to build oversight into an agent’s lifecycle, from checks before it commits to pull request quality gates and logs that track changes by agent and model. Then Ignacio will give a live look at how the Quality Clouds engineering team uses agents, harnesses, and frameworks, and how Norma checks their work before it’s merged.
You’ll leave with a pipeline checklist you can use with your tools. Want to try Norma? Explore the free version at norma.qualityclouds.com. Come prepared, and you may also get access to the PRO version through the end of the year.
AI can write the code and the tests in one go. But how do you know the tests are any good, and what would convince you to merge the change?
Jakub Marchwicki takes a fresh look at the testing pyramid through the lens of microservices and agentic coding. Drawing on a year of hands-on experience as an individual contributor, he’ll explore what to ask coding agents to test, how to judge their work, and whether the old testing rules still hold up.
Join Barcelona JUG for an evening of ideas, conversation, and a chance to challenge your own AI-assisted coding habits.