Dynatrace’s Alois Reitbauer explains why feature flags are the key safety mechanism for agentic AI in this episode of The New Stack Makers podcast.

The New Stack is a publication covering trends and technologies in cloud-native development, DevOps, and software delivery. Developers can learn about containerization, Kubernetes, and cloud computing, as well as explore topics such as microservices architecture, serverless computing, and continuous integration/continuous delivery (CI/CD) pipelines.

The New Stack

Feature flags can serve as safety mechanisms for agentic AI in production environments. Rather than allowing AI agents to modify source code directly, they should change configurations through feature flags, providing controlled autonomy within defined guardrails. Dynatrace's acquisition of DevCycle positions feature management as a key control layer, enabling AI to make faster decisions while maintaining safety constraints. This approach gives organizations confidence to adopt autonomous operations without surrendering full control to AI agents.

The reason AI agents shouldn’t touch your source code — and what they should do instead