Worker Versioning is now in public preview, letting you deploy workflow changes safely without breaking running instances. This solves a critical operational friction point.

Deploy workflow changes on your schedule without breaking in-flight executions, enabling true continuous deployment for stateful workloads.
Signal analysis
Temporal workflows are designed to run for days, weeks, or months. Until now, deploying code changes to Workers risked breaking active workflow executions if the new code couldn't understand the old execution state. Worker Versioning removes this constraint by letting you run multiple versions of your Worker code simultaneously, with Temporal routing work intelligently between them.
This is operationally significant because it eliminates the false choice between: (a) deploying immediately and risking in-flight workflow failures, or (b) waiting for all workflows to complete before updating. You can now deploy on your schedule, not theirs.
Worker Versioning relies on explicit version declarations in your Worker code. You mark which activities and workflows have changed, and Temporal uses that metadata to decide routing. The declarative approach means you're not fighting versioning hell—you're explicitly saying what's new and what's compatible.
For most teams, this means: tag your Worker with a version number, declare which activities have new signatures, deploy to your existing infrastructure, and let Temporal handle the traffic split. Rollbacks become trivial—just deploy an older version alongside the new one.
The win is clear: you get true rolling deployments for stateful workloads without the operational complexity of maintaining separate clusters or manual traffic management. This is particularly valuable for mission-critical workflows (payment processing, order management, etc.) where downtime is unacceptable.
The trade-off is that versioning only works if you plan for it during development. Backward-incompatible changes to activity signatures still require compatibility bridges or orchestration on your side. Worker Versioning handles the hard problem (routing), but doesn't magically fix code incompatibility—you still need good API design practices.
If you're running Temporal in production with long-running workflows, this is worth immediate evaluation. Audit your current deployment process and identify workflows where rollout speed is constrained by completion time.
For teams still on self-hosted Temporal, this feature requires server-side support; verify your version is compatible before assuming you can use it. Cloud customers get this automatically with the public preview.
Best use cases
Open the scenarios below to see where this shift creates the clearest practical advantage.
One concise email with the releases, workflow changes, and AI dev moves worth paying attention to.
More updates in the same lane.
The new Neon MCP connector transforms how AI tools interact with browsers, enhancing real-time automation and productivity.
Phidata's latest update enhances automation with Fallback Models support, improving task management for developers and teams.
The latest WordPress update empowers users with plugins and Global Styles on every paid plan, greatly enhancing customization options.