WigertzTech
All projects
Serverpod Logo
Serverpod2026, 5 months

Serverpod

The full backend stack for Flutter and Dart, deployed in one command

Cloud / DevOps / Developer Tools

Platform
Managed Dart hosting
Cloud
GCP / GKE
Engagement
5 months
Serverpod Logo

Serverpod is the open-source Dart backend for Flutter teams, and Serverpod Cloud hosts it, deployed in one command with no Kubernetes for the customer to run. Over five months I owned platform and infrastructure work on it and shipped product features in the platform itself. That covered the private link between the platform and the cluster running customer backends, the pipeline every infrastructure change goes through, and file storage, a feature customers now pay for.

01

The platform reaches every customer backend without touching the public internet

The service that manages customer backends runs in its own control cluster, and it has to call into the shared, multi-tenant cluster that hosts every customer's backend. I designed a private, mutually authenticated connection between the two, with certificates the platform issues itself, so that traffic never touches the public internet. It went from first spike to every environment inside a week, with runbooks for the teardown order that nobody gets right the first time.

Stack
Private Service Connect, cert-manager, Gateway API
Shipped as
Helm + Terraform
Rollout
Under a week

02

A deployment pipeline built as a product, not a folder of workflows

Every infrastructure change goes through one reusable pipeline that plans, releases and applies it, the same way for every module. An end-to-end test suite runs the whole chain against real modules and gates every merge. A stale plan is caught rather than applied, and no change is ever applied from inside a pull request. Every module in the infrastructure repo moved onto it in about a week.

Stack
Terragrunt, GitHub Actions
Migration
About a week

03

Live health for every customer backend, and new versions rolled out as canaries

The platform can now report each customer backend's health live. It is defined once as a shared contract, served from inside the tenant cluster and streamed into the product. The same cross-repo work taught the operator that runs customer backends to roll out new versions gradually, and took the platform's configuration from an empty repository to one where every change is a pull request showing exactly what each cluster will get.

Languages
Go, Dart, HCL
Repos
6, two bootstrapped
Stack
OpenAPI, Flagger, Flux

04

The same person who wires the cluster ships the product

File storage is a customer-facing feature with a price list, and I built all of it, from the cloud permissions to the product code that ties each bucket to its backend from creation to deletion. Then I made it a product. Usage is metered into billing, plan limits are enforced, every backend gets private and public storage by default, and revoked access, custom domains and rate limits are all handled. Platform engineers do not usually write the product, and product engineers do not usually hold the cloud permissions. Owning both is what turns a feature like this into weeks instead of a quarter of coordination.

Scope
IAM to billing
Per backend
Private + public storage

What I owned

Principal Engineer

  • Private link from the control cluster to the tenant cluster
  • Reusable infrastructure pipeline gated by end-to-end tests
  • Every infrastructure module migrated onto it
  • Live health reporting for customer backends
  • Canary rollouts in the tenant operator
  • GitOps for the platform layer, from an empty repo
  • File storage end to end, permissions to billing

GCP / GKE / Kubernetes / Go / Dart / Serverpod / Terraform / Terragrunt / Helm / Flux / Flagger / cert-manager / Gateway API / Cloud Storage / Cloud Monitoring / GitHub Actions

Related work

All projects
Pugstorm DevOps Platform Dashboard

Zero idle compute cost

Pugstorm

Automated builds and Steam releases for the studio behind Core Keeper

Principal Engineer

GitHub Actions / AWS / EC2 / S3 +6

Lynes Logo

Zero to self-service CI/CD

Lynes

From manual deploys to developer self-service

DevOps Architect & Engineer

Kubernetes / AWS EKS / Terraform / Terragrunt +9

Volvo Cars Logo

100+ apps across 4 regions

Volvo Cars

Platform engineering and SRE for a global developer platform

Principal Engineer

Kubernetes / AWS / Azure / Terraform +9

Next step

Tell me what you're building

Send me a short description of the problem and I'll tell you whether I'm the right fit, what it would cost, and how long it would take. No pitch deck.

Email meCall +46 70 753 73 78
Email
elinor@wigertztech.se
Phone
+46 70 753 73 78
LinkedIn
elinorwigertz