Compare
Every hosting option trades control for effort somewhere. This is an honest map of where ShipSanely sits next to a traditional PaaS, a Kubernetes cluster and raw virtual machines.
| ShipSanely | Traditional PaaS | Kubernetes | Raw VMs | |
|---|---|---|---|---|
| Setup input | Your repository | Dockerfile / buildpack + platform config | Manifests, Helm charts, cluster | Everything, by hand |
| Infrastructure definition | Derived from code by the agent | Written by you per service | Written and maintained by you | Written and maintained by you |
| Database & cache | Provisioned & wired automatically, private | Add-ons, configured manually | Operators or external services | Installed and secured by you |
| Verification before live | Built in: probes must pass | Build success only | Readiness probes if you write them | Whatever you script |
| Monitoring | Included | Separate tooling | Separate stack (metrics, logging, alerting) | Separate stack |
| Incident response | Agent diagnoses & repairs | Human on-call | Human on-call | Human on-call |
| Scaling | Agent-driven within your bounds | Manual dial or basic autoscale | HPA + your tuning | Manual |
| Lock-in | None — standard PG, Redis, containers | Platform SDKs / config | Low, but heavy operational cost | None |
| Pricing | Two meters: agent + infrastructure | Bundled, add-ons extra | Cluster + tooling + people | VMs + people |
| Best for | Teams who want an outcome, not a console | Simple apps with an ops-minded dev | Large orgs with platform teams | Specialists who want total control |
If you need a specific cloud region for compliance that we do not yet offer, run GPU workloads, or depend on a runtime outside our supported set, a general-purpose cloud will serve you better today. We would rather say so than onboard a project the agent cannot run well.
You have a web application with a database and a cache, a small team, and better things to do than write infrastructure config and carry a pager. That is the case the agent was built for.
Early access is open for a limited number of teams. Tell us about your stack and we will reply with next steps.