Systems | Development | Analytics | API | Testing

Every CIO has a slide deck full of AI pilots by now. A chatbot that answers FAQs. A copilot that drafts emails. An agent that summarizes meetings and files the notes nobody reads. The demos get applause in the boardroom, the budget gets approved for “phase two,” and then largely nothing happens. The pilot quietly lives on in a sandbox, forever 80% done, forever six weeks from launch. If that sounds familiar, you are not behind. You are, statistically, in the majority.

Every team we talk to has a running list of questions they wish they could get fast, reliable answers to. What changed in our performance last month and why. Which clients are showing the early signs of churn. Which channels are actually pulling weight and which ones are quietly burning budget. The pull toward AI for this kind of work is obvious. The answers should be a question away.

Data center teams are skilled at solving familiar problems such as storage outages, missed forecasts, and late refresh cycles. These are known quantities. Teams have playbooks for them. But 2026 has brought a different kind of pressure. After years of enterprise AI investment concentrated almost entirely on model training, the industry has crossed a threshold: the workload that now defines AI infrastructure isn’t building models. It’s running them. Continuously. At scale. Every day.

With the release of Xray Cloud 15.0.0, Xray expands its latest AI capabilities with the introduction of AI Test Prioritization, joining two recently released features: Xray's Rovo Test Plan Summarizer and AI-generated Manual Scripts for Test Case Designer, introduced in Xray Cloud 14.0.0. Testing is rarely just about executing test cases. Teams need to understand where risk exists, how testing is progressing, and whether a release is ready to move forward.

Webinar recap: Andy Cotgreave (co-founder, How to Speak Data) and Francois Lopitaux (SVP of Product, ThoughtSpot) on hallucinations, accountability, and what it actually takes to trust AI with your data.

You will still need to solve many last-mile problems to make them production-ready. We can help.

There's a well-worn path to resumable AI chat streams: find the Vercel SDK docs, implement Redis-backed replay, and ship it. For many products, that's the right call. The challenge arises when the product goes further than that. AI customer support tools that handle complex queries over 30-plus seconds. Agents that keep working while the user switches from their laptop to their phone.

In the first experiment, I wanted a baseline: if an AI coding agent gets the same production signal a human would get, can it fix bugs in a codebase it has never seen? Yes, but only when I gave it better context. With only an alert, the agent passed 51% of the runtime tests. When I added captured traffic, the actual request and response for the failing call, it climbed to 77%. This post is the second pass.