Dynamic Workflows: Enabling Per-Tenant Durable Execution at Scale

Cloudflare’s platform has evolved from a direct-to-developer tool to a rich ecosystem where platforms can build multi-tenant applications. With the introduction of Dynamic Workers, Durable Object Facets, and Artifacts, they’ve enabled dynamic compute, storage, and source control. Now, Dynamic Workflows bridges the gap between durable execution and dynamic deployment, allowing platforms to run custom workflows per tenant or agent. Here’s everything you need to know about this breakthrough.

How did Cloudflare Workers evolve from a direct-to-developer platform to enabling multi-tenant applications?

When Cloudflare launched Workers eight years ago, it was designed for individual developers to deploy code directly. Over time, they expanded the ecosystem so that platforms could build on Workers and also let their customers ship code through multi-tenant applications. Today, Workers supports diverse use cases: AI-generated implementations from user descriptions, multi-tenant SaaS where each customer’s business logic is unknown TypeScript at runtime, agents that create and run their own tools, and CI/CD products where each repo defines its own pipeline. This evolution laid the groundwork for dynamic deployment.

What are Dynamic Workers, Durable Object Facets, and Artifacts? How do they enable dynamic deployment?

In the Dynamic Workers open beta (shipped last month), platforms receive a clean primitive for compute: hand the Workers runtime some code at runtime, and get back an isolated, sandboxed Worker on the same machine in single-digit milliseconds. Durable Object Facets extends this to storage—each dynamically-loaded app can have its own on-demand SQLite database, with the platform acting as a supervisor. Artifacts does the same for source control, offering a Git-native, versioned filesystem that can be created by the tens of millions, one per agent, session, or tenant. Together, they form a stack for dynamic deployment across compute, storage, and source control.

What is Cloudflare Workflows, and what are its core capabilities?

Cloudflare Workflows is a durable execution engine that turns a run(event, step) function into a program where every step survives failures, can sleep for hours or days, wait for external events, and resume exactly where it left off when the isolate is recycled. It’s ideal for tasks that need to “keep going” past a single request: onboarding flows, video transcoding, multi-stage billing, and long-running agent loops. With Workflows V2, it supports up to 50,000 concurrent instances and 300 new instances per second per account, redesigned for the agentic era.

What limitation did Workflows have for multi-tenant scenarios?

Workflows always assumed that the workflow code is part of your deployment. Your wrangler.jsonc has a block that says “when the engine calls into WORKFLOWS, run the class called MyWorkflow.” One binding, one class per deploy. This works fine if you own all the code or run a traditional application. But it stops working when you want to let your customer ship their own workflow. For example, an app platform where the AI writes TypeScript for every tenant, a CI/CD product where each repo has its own pipeline, or an agent SDK where each agent writes its own durable plan. In all these cases, the workflow is different for every tenant, agent, or request—no single class to bind.

What problem does Dynamic Workflows solve?

Dynamic Workflows bridges the gap between durable execution and dynamic deployment. It solves the same shape of problem that Dynamic Workers solved for compute: allowing platforms to hand custom workflow code at runtime, rather than binding a single class at deploy time. Now, each tenant, agent, or request can have its own unique, durable workflow that survives failures, sleeps, waits, and resumes—just like traditional Workflows, but dynamically assigned. This enables platforms to offer per-tenant workflows without pre-deploying code, unlocking new possibilities for AI-driven agents, custom CI/CD pipelines, and multi-tenant SaaS.

How does Dynamic Workflows integrate with the existing dynamic deployment stack?

Dynamic Workflows completes the dynamic deployment trifecta alongside Dynamic Workers (compute), Durable Object Facets (storage), and Artifacts (source control). With these four primitives, platforms can give each tenant or agent an isolated, sandboxed Worker (Dynamic Workers), its own SQLite database (Durable Object Facets), a versioned filesystem (Artifacts), and now a durable execution workflow (Dynamic Workflows)—all spun up on demand. The platform sits in front as a supervisor, managing isolation and lifecycle. This holistic stack lets developers build truly dynamic, multi-tenant applications where every component is per-tenant, from compute to storage to orchestration.