Core service
Make your cloud infrastructure predictable
When Kubernetes, infrastructure code and the delivery pipeline are set up separately, every team finds its own way. A shared platform takes the infrastructure work off your teams.
- Duration
- 2 to 6 months
- Model
- Build and hand over
- Output
- Reference platform
Signs a platform is missing
The need for a platform is rarely noticed as an infrastructure problem. It shows up as delivery slowing down.
- Every team writes its own deployment scripts, and they all differ.
- Getting a new service into production takes days.
- Environments differ; what works in test does not work in production.
- There is a Kubernetes cluster, but nobody knows exactly how it is configured.
- Infrastructure changes are made by hand and are not recorded.
- The cloud bill cannot be broken down by team.
A platform is designed like a product
A good platform is a surface where product teams can ship services without thinking about infrastructure. That surface does not appear by itself; it is designed like a product whose users are your internal teams.
Everything is built with infrastructure as code, so every change can be reviewed, rolled back and reproduced. Networking, secret management and observability are part of the platform from the start, not added later.
The choice between managed cloud services and self-hosted infrastructure depends on cost, team capacity and data residency requirements. We compare the options for you in numbers.
Scope
Included
- Kubernetes cluster architecture and setup
- All environments defined as infrastructure code (Terraform)
- Delivery pipeline and GitOps workflow
- Network layer and service-to-service security
- Secret management
- Metrics, logs and tracing infrastructure
- Cost tagging and per-team reporting
- Runbooks and incident response procedures
Not included
- Rewriting application code
- 24/7 on-call and operations service
- Hardware procurement
- Licence resale
How the build runs
- Weeks 1-2Discovery. Existing infrastructure, deployment methods and team needs are mapped. The boundaries of the target platform are set.
- Weeks 3-6Foundation. Network, identity, cluster and infrastructure code are built. Every environment is produced from the same definition.
- Weeks 7-10Delivery and observability. The delivery pipeline, observability and alerting rules are put in place. The first service moves onto the platform.
- Week 11 onwardsHandover. Remaining services move across, runbooks are written and your team runs the platform on its own.
What you receive
The platform is built in your accounts, on your infrastructure. Every definition lives in your repository.
Who this service is not for
- Teams running a single application on a single server. At that scale a platform adds more complexity than it removes.
- Organisations that want to hand infrastructure entirely to an outside provider. This service aims to hand the platform over to your team.
- Organisations that have already made the platform decision and only want build labour. We question the decision itself as well.
Frequently asked questions
Do we really need Kubernetes?
Not always. For teams running a handful of services, managed container services or simpler options may fit better. We base our recommendation on team size and the number of services.
Which cloud providers do you work with?
AWS, Azure and Google Cloud, as well as self-hosted infrastructure. We do not resell any provider.
Whose account is the platform built in?
Yours. Cloud accounts, repositories and all access are opened in your organisation. Any access given to us is closed at the end of the project.
Will we need to change our existing applications?
Usually small changes are enough, such as moving configuration into environment variables and adding health checks. We settle the scope during the discovery weeks.
What happens after the build?
The goal is for your own team to run the platform. A period of advisory support after handover can be planned separately.
Let’s review your infrastructure together
In thirty minutes you walk us through your current setup; we tell you what a platform would give you and where to begin.