Systems | Development | Analytics | API | Testing

AI agents can write tests, but those tests can repeat the same assumptions that led to the code. In this walkthrough, I use Cursor and proxymock to replay real application traffic against a local Go app, compare requests and responses, and validate code changes with real data.

Rovo, Atlassian’s AI assistant, can help you generate test cases directly inside Jira through the SmartBear Zephyr Agent for Rovo. This demo offers a practical look at AI-powered testing with Rovo in Jira, from requirements to reviewed test cases, all within the Jira experience, without switching tools.

When a performance test run goes wrong, the real work begins, which means hours of manual analysis, digging through metrics, and trying to prioritize what to fix first. Agentic Performance Testing (APT) in NeoLoad changes that. In this demo, see how APT's specialized AI agents automatically analyze a failed test complete with a full, stakeholder-ready report, surfacing an executive summary, SLA compliance breakdown, trend analysis. critical findings, a prioritized action plan, and more, all without leaving NeoLoad.

This session takes a practical look inside LLM context windows and token consumption, exploring what happens when context enters a model - from tokenization, embeddings, attention, QKV, prefill, and decode to KV caching. It also examines how context windows are allocated and why simply increasing context length doesn’t always lead to better model performance.

Most AI pilots never make it past the demo phase. AWS Machine Learning Lead Praveen Jayakumar has seen plenty of promising AI projects get stuck between a successful demo and production. Teams often define what success looks like without deciding what failure looks like, leaving underperforming projects alive long after they should have been shut down. As Praveen puts it, they become “zombie” AI projects.