Is AI ready for DevOps?

In the inaugural episode of "Agentic DevOps," Bret Fisher interviews Nirmal Mehta at KubeCon 2025 about the emerging concept of AI Agents in DevOps and platform engineering. They define AI Agents as LLMs with the ability to execute commands via tools, contrasting this with current code editors. The discussion covers the Model Context Protocol (MCP) as a standardization effort for defining these tools and highlights the potential paradigm shift towards AI-driven automation in infrastructure management. They address concerns about introducing non-determinism into traditionally deterministic systems and explore the possible business pressures driving the adoption of AI agents, including potential workforce reductions and the need for increased velocity in DevOps. The episode concludes by touching on the emergence of AI gateways and the broader implications for the Kubernetes ecosystem and cloud-native technologies.