Mike Vizard interviews Loreli Cadapan to unpack how MCP, an AI-powered protocol, is redefining DevOps by integrating large language models into CI/CD and software delivery workflows.

How MCP Is Shaping the Future of DevOps Processes

Author: Mike Vizard
Interviewee: Loreli Cadapan, VP of Product at CloudBees

Introduction

The Model Context Protocol (MCP), pioneered by Anthropic, is drawing attention in the DevOps world. Loreli Cadapan from CloudBees explains how MCP can reinvent the foundation for enterprise DevOps workflows by acting as a control plane that connects large language models (LLMs) to CI/CD and related tools.

What Is MCP?

MCP bridges LLMs and DevOps environments.
It operates as a kind of control plane, able to interact with various CI/CD systems and testing frameworks.
The protocol can pull real-time data about pipelines, vulnerabilities, and orchestrate workflow tasks.
By reducing context switching and seamlessly integrating with developer tools, MCP aims to accelerate productivity and improve governance in software delivery.

How MCP Changes DevOps

MCP servers are compared to factory floor managers: they coordinate AI agents to perform tasks like diagnosing build failures or recommending tests.
The efficacy of these AI agents depends on the quality of their training data and the feedback loop for their outputs.
Adoption varies: some organizations leverage co-pilot style assistance (keeping humans in the loop), while others explore more autonomous workflows.
Guardrails and policies are necessary to ensure secure, predictable, and reliable outcomes.

Impact on the Software Delivery Lifecycle

MCP and AI agents may change not only coding and pipelines, but also product management, requirements drafting, and validation.
As AI-generated code and automated commits become widespread, downstream bottlenecks shift to reviewers and pipelines.
Teams are advised to stay educated, experiment carefully, and always validate AI outputs.

The Broader Future

The rise of MCP and AI-powered DevOps points to a future where software delivery is highly automated and AI-embedded at every stage.
Success hinges on updating pipelines, governance practices, and team workflows alongside this rapid evolution.

Key Takeaways

MCP facilitates intelligent automation for DevOps by linking LLMs to enterprise workflows.
AI agents can significantly raise productivity in CI/CD, but require careful setup and oversight.
Teams should be proactive and critical in their adoption of AI in DevOps.

For more details, watch the Techstrong Gang Interview.

Original article published on DevOps.com.

This post appeared first on “DevOps Blog”. Read the entire article here