--- title: "In The Age Of AI, Dev Workloads Are Cloud Workloads - Blog - Coder" description: "By migrating dev workloads to AWS via Coder, our joint customers centralize and govern the dev experience for both humans and agents." image: "https://www.datocms-assets.com/19109/1771941280-blog_coder-aws.png?fit=clip&fm=webp&w=800" canonical: "https://coder.com/blog/in-the-age-of-ai-dev-workloads-are-cloud-workloads" --- May 22 20264 min read # In The Age Of AI, Dev Workloads Are Cloud Workloads [Tim Cleary](https://coder.com/blog/author/tim-cleary) Share this article When [the world's leading AI safety company](https://coder.com/blog/inside-anthropics-ai-first-development) runs two to four Claude Code agents per developer on cloud workspaces, that's a dev workload running like a production workload. When [a government software group](https://coder.com/success-stories/us-department-of-defense-software-development-group) pulls source code off thousands of laptops and centralizes it inside AWS GovCloud, same thing. And when [a global fintech with 15,000 engineers](https://coder.com/success-stories/financial-services) stands up governed AI development infrastructure to satisfy compliance teams without slowing down its developers, same thing again. Customer stories like these are what earned Coder the [Amazon Web Services (AWS) Migration and Modernization Competency status](https://coder.com/blog/coder-achieves-aws-migration-and-modernization-competency-status) for [AWS Partners](https://coder.com/partners/aws). It's a recognition of the value we've created with customers and the work we've put into our products. And while every competency we've earned with AWS, in [DevOps, AI](https://coder.com/blog/coder-achieves-aws-generative-ai-and-devops-competencies), and [Government](https://coder.com/blog/coder-listed-in-aws-icmp-for-the-us-federal-government), has been a meaningful milestone, this one is particularly special. ![Coder's AWS Partner competencies](https://www.datocms-assets.com/19109/1779417944-aws-competency.png) Historically, [the Migration and Modernization competency](https://aws.amazon.com/migration/partner-solutions/) has been reserved for partners with a proven track record of moving production workloads from on-premises infrastructure to AWS. Think a legacy financial system getting lifted off local servers, or a monolith getting refactored into cloud-native architecture for better scalability and reliability. AWS cares a lot about getting this right and has invested heavily in the partner ecosystem to drive it forward. The Migration Acceleration Program (MAP) is the most established and successful partner motion in their catalog. Dev workloads, where software engineers and data scientists build and test the code that eventually ships, were for the most part left out of that conversation. They lived on laptops or inside VDI. Occasionally, an enterprising engineer would spin up dev containers in a VM and then suffer through the infrastructure overhead required to keep them running. That was a defensible model last year. It is not a defensible model now. In the last 12 months, vibe coding has caught fire in the enterprise, the SDLC is becoming the AI-DLC, and agents are becoming top committers at both bleeding-edge startups and dominant enterprises. AI coding assistant adoption inside engineering organizations has reached 91%, according to DX, and developers expect AI-authored code to grow from 42% of their commits today to 65% by 2027. The load that puts on a developer's environment, in compute, in security exposure, and in governance, isn't something a laptop or a VDI session was designed to handle. A distributed approach to dev might have been good enough last year. It cannot keep up with the demands of this one. Scalability, high availability, and centralized governance have always been requirements for production workloads. They are now just as relevant for dev. Enter [Coder Workspaces](https://coder.com/solutions/workspaces), [AI Governance](https://coder.com/solutions/ai-governance), and [Agents](https://coder.com/solutions/agents). By migrating dev workloads to AWS via Coder, our joint customers centralize and govern the dev experience for both humans and agents. Workspaces spin up in seconds. Compute scales vertically and horizontally on demand, just like prod. Identity, access, and audit are properties of the platform, not afterthoughts bolted on through endpoint policy. Together, [Coder and AWS](https://coder.com/partners/aws) give transformation leaders the leverage to scale AI assisted software development with control, speed, cost efficiency, and flexibility. In short, when it comes to AI, dev workloads are now cloud workloads. The implications of that shift are significant. Platform engineering teams have traditionally owned how compute and infrastructure get delivered to production. They now own the same job for development. Source code moves off endpoints and into governed AWS environments. AI tool usage becomes observable and auditable rather than shadowy. Onboarding shrinks from weeks to minutes. And the modernization conversation gets a new front door: the work of refactoring and re-platforming legacy systems happens inside cloud-native dev environments that mirror the target state, before any production workload moves. That's the story behind the competency. It's also the story we expect to keep telling as more enterprises wake up to the fact that the build side of the SDLC has the same operational requirements as the run side, and starts treating it that way. Learn more about how enterprises are using [Coder on AWS](https://coder.com/partners/aws). ### Subscribe to our newsletter Want to stay up to date on all things Coder? Subscribe to our monthly newsletter for the latest articles, workshops, events, and announcements.