
Deploy SoduWork
Self-hosted Hub where teams and their AI agents share work
Just deployed
/var/lib/postgresql/data
SoduWork
Just deployed
/data
Deploy and Host SoduWork on Railway
SoduWork is an open-source, self-hosted engine for teams and their AI agents. A central Hub keeps issues, plugins, chat history, a shared library and model settings, while the SoduWork desktop client runs agents (Pi, Codex, Claude Code and more) on each person's machine and reports back to the Hub.
About Hosting SoduWork
This template runs the SoduWork Hub, built from the official ops/self-host/Dockerfile, next to a Railway PostgreSQL database. The Hub serves the admin console and the API on one public HTTPS domain. Uploads, the library, workspaces and repository mirrors are kept on a Railway volume mounted at /data, so they survive redeploys. The admin password, signing secret and connector encryption key are generated for you at deploy time. After the first deploy, open the domain, sign in as admin with the ADMIN_PASS value from the SoduWork service's Variables tab, add a model provider under Model catalog, then point the desktop clients of your team at the same address.
Common Use Cases
- A shared Hub for a team that runs coding and office agents on their own computers
- Central model catalog, plugins and permissions for many agent runtimes
- A self-hosted alternative to hosted multi-agent workspaces, with data in your own database
Dependencies for SoduWork Hosting
- PostgreSQL (included in this template)
- At least one model provider key, added later in the admin console
Deployment Dependencies
- SoduWork repository: https://github.com/mdiri-ai/SoduWork
- Self-hosting guide: https://dev.soduwork.ai/docs/deploy
Implementation Details
The Hub runs with SODUWORK_ENV=production, so it refuses to start with weak secrets, SQLite or CORS_ORIGINS=*. HUB_PUBLIC_URL and CORS_ORIGINS follow the Railway domain automatically. Execution on the Hub itself is off (HUB_LOCAL_EXECUTION_ENABLED=false): agents run in the desktop clients. The start command links the Hub's data folders to the /data volume before starting uvicorn on port 8989.
Why Deploy SoduWork on Railway?
Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it.
By deploying SoduWork on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway.
Template Content
SoduWork
mdiri-ai/SoduWork
