Sam and Dave discuss that while AI may offer useful tools and incremental improvements in DevOps, it will not fundamentally change its core focus on culture, collaboration, and continuous delivery. They caution against over-reliance on AI for automation and emphasize that human involvement remains essential to maintaining the true essence of DevOps.
In this discussion on the Modern Software Engineering channel, Sam and Dave explore whether AI will fundamentally change DevOps. Both agree that while AI might introduce some useful tools and incremental improvements, it is unlikely to transform DevOps at its core. Dave begins by emphasizing that DevOps is fundamentally about culture and collaboration among people with different expertise working together to deliver software effectively. Sam adds that DevOps is part of the broader concept of continuous delivery, which focuses on continuously delivering value to users by bridging the gap between development and operations teams.
They discuss various models of understanding DevOps, including the CALMS model—Culture, Automation, Lean, Measurement, and Sharing. They agree that AI’s impact on culture and sharing is limited, as these aspects rely heavily on human interaction and collaboration. Automation is another area where AI’s role is constrained because automation requires deterministic, reproducible, and auditable processes, which current AI models, especially large language models, are not well-suited to provide. AI might assist in triggering scripts or helping write code but cannot replace the need for precise and reliable automation pipelines.
Regarding measurement and observability, both acknowledge some potential for AI to help identify anomalies or patterns in production systems. However, they caution that such AI-driven insights are unlikely to be fully autonomous or universally applicable anytime soon. The complexity and context-specific nature of software systems mean that AI tools would require significant tuning and human oversight. They also express concern that reliance on AI might reduce valuable human conversations and collaboration, which are essential to the DevOps culture.
The conversation also touches on the dilution of the term “DevOps” in the industry, with many people misunderstanding or misusing it. Titles like DevOps engineer, site reliability engineer, and platform engineer are often conflated or used without clear distinctions, sometimes driven by corporate trends or vendor marketing rather than genuine role definitions. This trend contributes to confusion about what DevOps truly means and its cultural significance.
In summary, Sam and Dave conclude that AI will not change DevOps forever but may provide helpful tools around the edges. The essence of DevOps remains rooted in culture, collaboration, and continuous delivery of value. They advise caution in over-relying on AI for automation and emphasize the importance of human involvement in the process. The discussion ends with an invitation for viewers to share their thoughts and questions about the future of DevOps and AI.