Every feature on every tier.
You move up a tier for scale, not to unlock a capability. Start free with a real application in production.
Free forever plan, no commitment, in your own cloud account.
Free forever
£0/mo
No commitment
Run a real app in production, on us.
Every feature, sized for a first team service.
- 1 production Kubernetes cluster
- Plan editor and tester agents
- Wayfinder Assistant in the UI, the CLI and your IDE
- Catalogue, policies, RBAC, SSO and audit
- AWS, Azure or Google Cloud
- 1M AI credits a month
- 3 cloud and 3 Kubernetes components, 2 environments, 1 seat
Starter
£399/mo
£3,990 a year
First real team deployment
Small teams, 2 to 5 engineers.
- 1 cluster
- 5 cloud components
- 5 Kubernetes components
- 5 environments
- 100M AI credits a month
- 3 users
ProfessionalMost popular
£2,499/mo
£24,990 a year
Scale across teams and clouds
Growing companies, 5 to 50 engineers and up.
- 10 clusters
- 150 cloud components
- 75 Kubernetes components
- Unlimited environments
- 500M AI credits a month
- 25 users
Enterprise
Custom
Annual contract
Unlimited scale, SLA and support
Large organisations.
- Unlimited clusters
- Unlimited cloud components
- Unlimited Kubernetes components
- Unlimited environments
- 1bn AI credits a month
- Unlimited users
Every tier includes the catalogue and repo templates, cloud resource and Kubernetes resource plans, workflows, agents and the Assistant, integrations with webhooks, web APIs and MCP, DNS management, SSO, RBAC and audit, the CLI, the UI and the API, and AWS, Azure and Google Cloud.
Questions
How is this different from a developer portal?
A portal catalogues the infrastructure you already have. Wayfinder provisions and runs it. Developers describe an app in one file and deploy it to your own cloud, agents and workflows do the operational work, and the platform team sets what each team can reach.
Can we bring our own Terraform modules and Helm charts?
Yes. Any Terraform or OpenTofu module becomes a cloud resource plan and any Helm chart becomes a Kubernetes resource plan. An agent drafts the plan from the module, tests it against your real cloud, and you fix the values that matter before publishing it to the catalogue.
Can we define our own agents, workflows and integrations?
Yes, and that is the point. An agent is a prompt, the tools it may call, the skills it can read, the shape of the answer it returns and the identity it runs as. A skill is written knowledge, with its own files and tools, that you author once and let the right agents read. A workflow is a trigger and a set of tasks. An integration packages inbound webhooks, outbound web API operations and an MCP server. All three are configuration, and you choose which teams may write them.
How do the agents work, and can we trust them?
Each agent runs as its own service account with the roles you grant it, and can never hold more than the person who wrote it. Agents with a shell run in a sandbox on a node created for that run. Every tool call and every conversation is recorded and can be replayed.
Does Wayfinder replace the tools we already use?
No, it connects to them. Terraform and OpenTofu, Kubernetes, Helm, GitHub, Datadog, Grafana, SigNoz, PagerDuty, Jira and Slack are integrations today, and anything with a webhook, an API or an MCP server can be added.
Which clouds does it run on, and where does our data live?
AWS, Azure and Google Cloud, plus any Kubernetes cluster you connect. Wayfinder provisions into your own accounts, so the infrastructure and the data in it stay yours. Platform data is stored in the UK, and customer data is never used to train models.
Which AI provider is used, and can we use our own key?
Set the provider for your tenant and supply your own key. Agent usage is metered in AI credits per turn.
Will we be locked in?
Your infrastructure stays defined as code in your own repositories and runs in your own cloud accounts. If you stop using Wayfinder, the modules, charts and resources are still there.
What is an AI credit?
Credits are how agent work is metered: drafting and testing plans, writing a manifest from a repository, running a deployment, and conversations with the Assistant in the UI, the CLI or your IDE. Credits reset each month, and every tier includes enough to do real work.
Does the free plan need a card?
Yes. Free Forever costs nothing and has no end date, but it needs a card on file.
Is annual billing available?
Yes. Starter and Professional are cheaper paid annually, shown under each price. Enterprise is an annual contract.
How quickly can we get started?
Start free and deploy a real service to your own cloud the same day. There is nothing to install: connect a cloud account and a repository and go.
Something not answered here? Ask us.
Start deploying to your own cloud today.
Start free in your own cloud account, or book a demo with the team.