Home · Services
Services.
Every service is available as a one-off build, a fixed-scope engagement, or an ongoing retainer. Pick the one that matches your situation — each page lets you watch it handle a bad day.
Cloud Platforms
AWS
End-to-end AWS architecture & ops
What happens when…
- An AZ goes dark
- Someone breaks a rule
- New team, new account
Google Cloud
GKE, Cloud Run, BigQuery
What happens when…
- Traffic bursts from zero
- Nightly data pipeline
- Link to AWS drops
Hybrid & Multi
AWS ↔ Hetzner ↔ on-prem
What happens when…
- Direct Connect fails
- Hetzner runs out of room
- Malware hits a server
Compute & Containers
EC2 / VMs
Right-sizing, autoscaling, AMIs
What happens when…
- Traffic spikes
- An instance goes bad
- Security patch lands
ECS Fargate
Production patterns at scale
What happens when…
- Traffic spikes
- A bad deploy goes out
- A task crashes
Kubernetes & EKS
GKE, EKS, self-managed
What happens when…
- A node dies
- Pods can't fit
- Cluster upgrade
Portainer & Docker
Small-team container ops
What happens when…
- Ship an update
- A container crashes
- Roll back a release
Data
Relational & SQL
PostgreSQL, MySQL, RDS, Aurora — managed, tuned, secured
What happens when…
- Primary fails
- Read traffic spikes
- Someone drops a table
In-Memory & Cache
Redis, ElastiCache — caching, queues, pub/sub
What happens when…
- Cache starts cold
- Cache node fails
- Hot key expires
NoSQL
DynamoDB & MongoDB — modelling & migrations
What happens when…
- One partition runs hot
- New access pattern
- Feed search & analytics
Data Lake & Storage
S3, S3 Tables, Apache Iceberg, Athena, Glue
What happens when…
- New data arrives
- A query scans everything
- A bad load lands
On-Prem & Hetzner
Hetzner Cloud
VPS, Dedicated, Load Balancers
What happens when…
- A server dies
- Data gets corrupted
- A disk fills up
Private Datacenter
Proxmox, OpenStack, KVM
What happens when…
- A host fails
- A host needs patching
- A disk fails
Networking (Mikrotik / Cisco)
Routing, VPN, BGP
What happens when…
- Direct Connect fails
- A new office opens
- A bad route change
DevOps & Automation
Terraform / IaC
Modules, Terragrunt, drift
What happens when…
- Someone edits prod by hand
- Spin up staging
- A plan deletes the DB
CI/CD Pipelines
GitHub Actions, GitLab CI
What happens when…
- Ten PRs land at once
- A test fails
- Release to production
GitOps & ArgoCD
Declarative K8s deployments
What happens when…
- Ship a new version
- Someone edits live
- Bad release — roll back
Observability
CloudWatch, Grafana, Prometheus
What happens when…
- Checkout gets slow
- One fault, forty alerts
- Find one slow request