AI and DevOps Evolution: Insights from DevOpsDays Singapore 2025

Key Concepts

• CALMS framework (Culture, Automation, Lean, Measurement, Sharing)
• AI-driven tools in DevOps workflows
• Microservices and MCP (Microservices Communication Protocol)
• AI architectures and ecosystem integration
• AI Ops
• Data Scientist as an agent
• AI governance and ethics
• Prompt as an asset
• Skills transition in the age of AI
• Infrastructure as Code evolving to "Everything as Code"
• Business domain knowledge importance

DevOps Landscape and the Evolution of CALMS

Sergiu, organizer of DevOpsDays Singapore, discusses the evolution of DevOps practices in the ASEAN region. He highlights the CALMS framework (Culture, Automation, Lean, Measurement, Sharing) as a cornerstone of DevOps transformations. He proposes replacing "Automation" with "AI" in the CALMS framework, reflecting the increasing prevalence of AI-driven tools in building, launching, and monetizing products, as well as providing feedback loops. DevOpsDays Singapore on May 14th and 15th included technologists, DevOps, data scientists, and AI specialists.

AI Architectures and Integration

Sergiu notes the increasing adoption of microservices and the MCP standard. He emphasizes how AI is enabling businesses to build solutions and integrate various parts of their business ecosystems (e.g., logistics, online banking) much faster, reducing development time from months/years to days/weeks. This involves a complete architectural revamp using AI for coding, deployment, and AI Ops. AI Ops is showing significant results after years of growth.

Data, Modeling, and the Role of AI

The discussion touches on the evolution of data science. Sergiu suggests that AI can act as a "Data Scientist agent," analyzing and summarizing data to provide insights to non-tech savvy individuals, enabling faster feedback loops and feature incorporation.

AI Governance and the "Prompt as an Asset"

AI governance is identified as a critical area, encompassing regulation, ethics, and asset management. A key point is the concept of a "prompt" as an asset, representing business logic expressed in natural language. Securing and protecting these prompts is crucial. The conversation in Open Spaces highlighted the evolution of what constitutes an asset: from binary to source code, and now to prompts.

Impact on IT Departments and Skill Transition

Sergiu predicts significant changes in IT departments over the next two years, similar to the impact of containers. He anticipates a condensation of roles, with AI handling translations and level 1/2 support tasks. Business users will be able to use prompts directly, with IT focusing on review, architecture, security, scaling, resiliency, and availability.

Skills for the Future

The discussion shifts to the skills needed to adapt to the AI-driven future. The concept of "code" is evolving, encompassing prompts and natural language. The need for traditional coders will decrease, while the importance of business domain knowledge and understanding customer needs will increase. Sergiu envisions more people becoming tech-savvy and efficiently using AI-driven tools. The operations aspect will also change, with AI enabling more efficient scaling and market share capture.

Notable Quotes

• "Container Wars" - Sergiu's term for the competitive landscape of container technologies around the time Pivotal was founded.
• "What if the Data Scientist could be an agent?" - Sergiu, highlighting AI's potential to automate data analysis.
• "Prompt as an asset" - Sergiu, emphasizing the value of natural language instructions for AI systems.

Technical Terms

• CALMS: Culture, Automation, Lean, Measurement, Sharing - A DevOps framework.
• MCP: Microservices Communication Protocol - A standard for microservices communication.
• AI Ops: Artificial Intelligence for IT Operations - Using AI to automate and improve IT operations.
• Infrastructure as Code: Managing and provisioning infrastructure through code rather than manual processes.
• Compliance as Code: Automating compliance checks and enforcement through code.

Synthesis/Conclusion

The conversation highlights the transformative impact of AI on DevOps practices, architectures, and skill requirements. The CALMS framework is being re-evaluated to incorporate AI, and the concept of "code" is expanding to include natural language prompts. AI is enabling faster development cycles, more efficient operations, and a shift in IT roles towards architecture, security, and scaling. Adapting to this future requires a focus on business domain knowledge, AI governance, and the ability to leverage AI-driven tools effectively. The prompt is now a valuable asset that needs to be secured.