VlbgWebDev: Autonomous AI & Guardrails

Houston, we have a Problem. Autonoumus AI got this. by Manuel Tscholl

What happens when AI not only detects that something is going wrong with a digital product, but also takes action on its own? How far can automation go today, and what new possibilities are emerging?
In this talk, we’ll show what’s already possible today and how Agentic AI could transform the maintenance of digital products in the future.

Beyond the Prompt: Agentic Software Development with Guardrails by Stephen Renwick

Agents can read, refactor, and generate code faster than we can review it—but we’re still responsible for making sure it’s fit for purpose. I’ll show how I use reusable skills, shared instructions across tools, and worklogs to make agent-assisted development more consistent and easier to follow. We’ll explore progressive disclosure—giving agents the context they need, when they need it—alongside tracking decisions across sessions and retaining control over what changes.