AI-assisted development is making it easier to build useful applications quickly. Getting an app running, however, leaves several questions unanswered: what can it access, how could it fail, and who is responsible when other people start depending on it?
Platforms can provide isolation, security checks, access controls, and runtime monitoring, but deciding what to enforce and where to draw the boundaries remains a practical challenge. The same controls that contain risk can also make experimentation harder.
How much should we trust code that has not been thoroughly reviewed? Which safeguards should the platform enforce by default? How should apps access secrets, internal services, and data? What needs to change before an experimental app can be considered for production?
This Birds of a Feather session brings together platform engineers, developers, SREs, and security practitioners to discuss these questions. VibeGrid, Roku’s environment for experimental and vibe-coded apps, will provide a starting point for participants to compare approaches and share their experiences.
The goal is to identify useful safeguards, understand the trade-offs between developer freedom and operational responsibility, and clarify the questions teams need to answer as prototypes become tools that people rely on.
{{ gettext('Login to leave a comment') }}
{{ gettext('Post a comment…') }}{{ errorMsg }}
{{ gettext('No comments posted yet') }}