AI Agents Accelerate Cloud Migration for On, Cutting Weeks to Days for Critical Services

On, the premium sportswear brand, recently announced a significant achievement in its cloud migration journey, utilizing AI agents in partnership with Google Cloud. The company successfully migrated 24 core services to Google Cloud, dramatically cutting the migration time from an estimated three months per service to just two weeks. This was accomplished by a small team of On's internal engineers, who executed 15 of these migrations completely in-house, with minimal planned downtime—under five minutes per service. The multi-agent AI architecture employed by Google Cloud divided the migration into specialized, automated steps, including codebase analysis, infrastructure configuration generation, translation pipeline creation, and testing validation. This development is highly significant for cloud and DevOps practitioners. It demonstrates a tangible, real-world application of AI in accelerating and streamlining one of the most challenging aspects of cloud adoption: migration. The ability to reduce migration timelines from months to weeks, coupled with minimal downtime and reduced reliance on external integrators, directly impacts project costs, resource allocation, and business continuity. For organizations struggling with the complexity and duration of moving legacy systems or expanding their cloud footprint, this agent-led approach offers a compelling blueprint. It empowers internal teams, allowing them to retain control and build in-house expertise, rather than outsourcing the entire process. The productivity gains are substantial, as engineers can offload repetitive and mechanical tasks to AI agents, focusing instead on higher-value architectural decisions and strategic oversight. This move by On and Google Cloud fits squarely within the broader trend of AI-driven automation permeating cloud and DevOps practices. The industry has been steadily moving towards greater automation to manage increasingly complex cloud environments. From infrastructure-as-code to CI/CD pipelines, the goal has always been to reduce manual effort and human error. The introduction of sophisticated AI agents takes this a step further, enabling intelligent automation that can understand, plan, and execute complex tasks like cloud migration. This is not merely about scripting; it's about agents performing tasks such as mapping inconsistent deployment patterns across multiple code repositories in hours, a task that would traditionally take days for human engineers. This trend is also reflected in other areas, such as the increasing focus on AI-assisted operations as a top modernization priority for enterprises. In practice, this means practitioners should begin exploring how AI agents can be integrated into their cloud migration strategies. This doesn't necessarily imply a complete handover to AI, but rather a strategic partnership where AI handles the heavy lifting of data replication, configuration scripting, and environment validation, while human engineers provide oversight and make critical decisions. Organizations should investigate platforms and tools that offer agentic capabilities for migration and consider pilot programs to understand the benefits and challenges within their specific contexts. It also highlights the growing importance of a unified data and compute foundation, as emphasized by Google Cloud, to fully leverage the power of agentic AI. The trade-off is a potential upfront investment in understanding and integrating these new AI tools, but the long-term gains in efficiency, speed, and reduced operational overhead appear to be substantial.