Vercel Adds Background Workers and Persistent Storage
The platform now supports long-running containers and Python, enabling more complex backend workloads for developers.
Updated on Oct. 15, 2026 in Data Centers
Live Poll
Do you prefer to host all your backend services on a single integrated development platform?
Vercel has launched Daemons in public beta, allowing developers to deploy containers with specific process counts per region. Additionally, the platform has made Python support and persistent storage via Drives generally available for all users.
Why it matters
These additions provide developers with the granular concurrency control and persistent storage state necessary to host asynchronous workloads and AI agent sandboxes. By integrating these capabilities directly into the Vercel ecosystem, the company expands its utility for complex backend processing.
Vercel Drives support persistent storage ranging from 1 GiB on Hobby plans to 16 TiB in maximum configurations. Writes are performed on local NVMe disks before streaming asynchronously to object storage to ensure persistence.
The players
Vercel
A cloud platform focused on frontend and serverless backend hosting, known for its integration with the Next.js framework.
The details
Daemons allow developers to manage background processes by specifying container counts per region via Git or the Vercel Command Line Interface (CLI). Python support enables native deployment for frameworks such as FastAPI, Flask, and Django, moving beyond previous language limitations. Drives provide persistent storage by acting as a mountable file system for agents in new sandboxes, bridging the gap between stateless serverless functions and long-lived state requirements.
Timeline
October 15, 2026: Vercel announced the suite of features during its Ship conference.
The Tech Race
This release positions Vercel to better compete with traditional container orchestration platforms by relaxing the stateless nature of serverless computing. It signals a shift toward supporting heavier backend AI agents that were previously difficult to maintain within standard function lifecycles.
Developers can immediately utilize Daemons and Python support to migrate asynchronous workloads or AI agents onto the Vercel control plane. These features are available across all Vercel plans, allowing teams to scale storage from 1 GiB up to 16 TiB based on their specific infrastructure requirements.
The takeaway
Vercel is effectively blurring the line between managed frontend hosting and full-stack backend infrastructure. Teams should monitor how these persistent storage integrations affect long-term cost models compared to traditional virtual private server solutions.
Further reading
For more on how infrastructure is evolving to support modern web applications, explore our Data Centers section.
Source note: This article includes information reported by The New Stack.
Live Poll
Do you prefer to host all your backend services on a single integrated development platform?
