Rumus Cloud runs your workloads on our own metal in Jakarta. Pick a maintained template and it is live in one step, or connect a GitHub repository and every push builds and deploys. No Kubernetes to operate, no YAML to memorise.
Free tier includes one service with 512 MB memory. No card required to start.
Every option ends in the same place: a container with a public address, a TLS certificate, logs and metrics.
PostgreSQL, Redis, n8n, WordPress, Directus, MinIO, Ghost, Metabase and others, pinned to a version and kept current by us. Choose one, set the environment variables, deploy.
Connect a repository and a branch. We detect the runtime, build an image on our builders, and start the container. Pushes to that branch redeploy automatically, with the previous image kept for rollback.
Point at any registry that speaks the OCI standard. We pull, run, and manage the lifecycle. Useful when your build already happens in your own CI.
The parts that usually take a week of glue work are part of the platform itself.
Certificates issued and renewed automatically for every service you publish.
New container starts and passes its health check before the old one stops.
Structured logs with retention, plus CPU, memory and request graphs per service.
Daily database snapshots kept for 14 days, restorable from the dashboard.
Services in one project reach each other by name, without touching the internet.
Secrets stored encrypted, versioned, and injected at start rather than baked in.
The last ten successful builds stay available, each one deployable again.
Priced per resource held, shown hourly, capped by a monthly ceiling you set.
Each template is a working configuration, not just an image reference: ports, volumes, health checks and required variables are already described.
| Template | Version pinned | Volume | Typical start |
|---|---|---|---|
postgresql | 16.4 | 20 GB | 38 s |
redis | 7.4 | 2 GB | 14 s |
n8n | 1.68 | 5 GB | 41 s |
wordpress | 6.7 | 10 GB | 52 s |
minio | RELEASE.2026-08 | 50 GB | 26 s |
directus | 11.3 | 10 GB | 46 s |
metabase | 0.52 | 5 GB | 63 s |
ghost | 5.9 | 5 GB | 35 s |
Running a platform is one half of the work. The other is teaching your team to build with the models. Our instructors run live cohorts on the tools we use in production.
Set up Claude Code against a real repository: repository context, review workflow, guardrails, and measuring what it saves.
Run the Hermes agent with Claude as the model, connect MCP servers, and expose workflows to a team without leaking keys.
From Console organisation and API keys to a multi-channel assistant that answers in WhatsApp, with usage tracking and cost control.
The platform itself uses the Claude API in production: deployment failure analysis, log summarisation, and an assistant that answers customer questions from our own documentation. Read how it is wired in the integration notes.
Incidents are published with a timeline, and each release is listed in the changelog. Current platform status: All systems operational