Systems | Development | Analytics | API | Testing

Every tech leader is talking about agentic development and the promise of dramatically speeding up software delivery. But as organizations begin scaling AI-generated code, many are running into the same challenge: their test data processes can't keep up.

The challenge for most data teams isn’t a lack of ideas—it’s the time it takes to turn those ideas into something usable. In this session, Steffen Bischoff, Chief Architect Data at Qlik, follows a single dataset from a core system through its entire journey to becoming a governed data product. You’ll see pipelines created by describing intent instead of writing code, versioned in Git, then curated, quality-checked, and documented with the help of specialized agents. From there, the data product is made available to the AI tool of your choice through the Qlik MCP Server.

Learn more: speedscale.com.

Now that AI coding tools have put development capabilities into more hands, prototypes are becoming business-critical applications almost overnight. Shanea Leven sees an opportunity for a new generation of builders, provided the infrastructure around their applications keeps pace. Shanea explains how organizations can give developers and new technical employees room to build while maintaining the standards required for enterprise software..

If an AI agent gives a business user the wrong number, and a consequential decision gets made off it, who's responsible? The data analyst who built the semantic layer? The platform? Or the business user who asked the question and acted on the answer?

Every AI agent you plug into your stack runs up a bill, one token at a time. Third-party systems, LLM calls, credit checks. In lower environments, that adds up fast, and sometimes those services aren't even available to test against.