- Recommended public repository name:
mlflow-tracking-server - Recommended product description: containerized MLflow tracking server deployment for PostgreSQL-backed metadata and optional S3 artifact storage.
- This repository is about deploying and operating MLflow itself, not building a model or agent runtime.
- Ground all documentation in the actual files and deployment pattern present in this repository.
- Do not invent internal infrastructure or hidden services.
- Treat PostgreSQL credentials, S3 credentials, Railway variables, and server URLs as private.
- Keep the README focused on why to self-host MLflow, how to deploy it, and how to use it with related projects.
- When referencing related repositories, describe them as examples or companion projects, not hard dependencies.
- containerized MLflow server based on
Dockerfile - PostgreSQL-backed metadata store
- optional S3-compatible artifact storage
- Railway deployment guidance
- companion guidance for using this server with agent evaluation and prompt optimization workflows
Safe to remove or avoid committing:
.envfiles- local MLflow state directories
- local SQLite artifacts
- one-off test outputs
Use caution around:
- deployment commands
- artifact storage configuration
- any example command that may imply real credentials