Far more information
Every config, every log and all the existing docs get read in full instead of skimmed. Decisions are based on the whole system, not on whatever there was time to look at.
AI-augmented engineering
Cloud architecture &engineering— from concept to production.
Corpoddon supports organisations in the design, migration and operation of their cloud infrastructure. European providers first (Scaleway, Hetzner, OVH, Clever Cloud), complemented by AWS or Azure when needed, together with Proxmox, on-premise and hybrid setups. Clients in France, Germany and across Europe.
I build infrastructure with AI agents, which makes things a lot faster. Nothing ships until I’ve read and tested it.
- Scaleway
- Hetzner
- OVHcloud
- Clever Cloud
- Proxmox
- Kubernetes
- Talos
- Terraform
- Ansible
- GitLab CI
- AWS
- Azure
01AI
AI does the heavy lifting. I stay in charge.
I use AI agents every day: to write Terraform, comb through a config or draft a runbook. It saves me a lot of time, and saves you money. But I read, test and understand everything before it reaches a client.
Why AI?
Because no human can read everything.
An infrastructure means thousands of lines of configuration, months of logs and documentation scattered all over the place. Nobody has time to read all of it carefully. An AI does. That is where it really makes a difference.
Fewer blind spots
A forgotten firewall rule, a bucket left public, a dependency hanging by a thread: AI explores the leads nobody would have thought to check. Then I verify everything it brings up.
Many more tests
When AI drafts a test, writing one more costs almost nothing. So we cover edge cases, failures and rollbacks instead of stopping at the scenario where everything goes right.
Infrastructure code, written faster
Agents produce a first draft of Terraform, Ansible or Kubernetes manifests from the spec. I review it and test it before it gets anywhere near production.
More thorough audits
An automated first pass over configs, IAM permissions and network rules catches what you miss in hour five of reading. Then I sort through it and decide what actually matters.
Docs that keep up
Runbooks, diagrams and reports are updated with every change, then reviewed. You don’t inherit a wiki nobody has touched in two years.
Your cloud bill, line by line
AI helps me dig through bills and metrics to find oversized instances, forgotten resources and workloads that could run on spot.
The rule
Nothing ships until I’ve reviewed it. And if something breaks, you call me, not a language model.
02Services
Twelve engineering practices, one office.
Each engagement starts with a written scope and ends with documented, reproducible infrastructure. No black boxes.
I
Approach
01
Design & Engineering
Reference architectures, technical specifications, capacity sizing and migration plans. Engineering deliverables you can hand to any operator.
AI-accelerated
02
R&D & Innovation
Research and development of innovative products and solutions. Prototyping, technology scouting, proof-of-concept engagements, and taking promising ideas from lab to production.
AI-accelerated
03
Eco-Design & Low-Carbon
Infrastructure choices grounded in sustainability. Workload placement on regions and providers served by low-carbon electricity grids, energy-efficient sizing, and a measurable CO₂ footprint for every deployment.
II
Infrastructure
04
Sovereignty & EU-First
Infrastructure decisions that keep data, code and operations inside Europe. Priority to European providers (Scaleway, Hetzner, OVH, Clever Cloud) and EU-resident regions on hyperscalers, with staged migration paths when data sovereignty becomes a requirement.
05
Cloud & Hybrid Architecture
Scalable, secure architectures with a European-first bias (Scaleway, Hetzner, OVH, Clever Cloud), opening to AWS or Azure when the use-case demands it. On-premise and hybrid topologies tailored to regulatory and cost constraints.
06
On-Prem & Bare-Metal
Custom on-premise and bare-metal solutions. They make the most sense for workloads with heavy egress, where cloud bandwidth pricing becomes the economic bottleneck. Proxmox VE clusters, VM and container orchestration, high-availability setups, backup strategies.
III
Operations
07
CI/CD & Automation
End-to-end CI/CD pipelines on GitLab and GitHub. Infrastructure as Code with Terraform and Ansible, Kubernetes, Docker, GitOps workflows and reproducible deployments.
AI-accelerated
08
Kubernetes · Off-Hyperscaler
Kubernetes clusters built outside the hyperscaler comfort zone: bare-metal, on-premise, or on European providers. Distribution selection (K3s, Talos, RKE2, vanilla k8s), lifecycle and upgrade strategies, storage and networking without AWS/Azure lock-in.
09
FinOps & Spot Leverage
Aggressive use of spot, preemptible and reserved instances across providers (AWS, GCP, Scaleway, Hetzner, OVH). Workload scheduling, budget guardrails and cost observability, so cloud bills stay predictable as infrastructure grows.
AI-accelerated
IV
Delivery & Audit
10
CDN & Edge Deployment
Global content delivery and edge computing. CDN architecture, cache strategies, serverless edge functions and end-to-end performance work.
11
Security & Compliance
Security baselines, identity and access management, network segmentation and GDPR / DSGVO compliance for cloud and hybrid setups.
AI-accelerated
12
Technical Consulting
Architecture reviews, technology selection, technical due diligence and second-opinion engagements for ongoing projects.
AI-accelerated
03Method
From first call to production, in four steps.
Scoping
You walk me through your context, in writing or on a call, and we agree on the scope and on what success will look like.
AII have your existing docs summarised so I show up to the first call having read everything.
Design
Target architecture, sizing, provider selection and the migration plan.
AISeveral options compared and costed before we decide.
Build
Infrastructure as Code, CI/CD pipelines, tests, then a migration in stages.
AIAI writes the first draft of the IaC and tests, I review every line.
Run
Monitoring, security, cost tracking, and a handover so your team can run it without me.
AIRunbooks and reports write themselves, and alerts arrive with their context.
04Tools
Small tools, open to everyone.
Spend enough time debugging networks and you end up building your own tools. Here are two you can use right now.
Web · API
ip.alo.is
Your public IP address, no frills. Plain text for your scripts, or JSON with country, city, ASN and network operator.
Your IP, live
$ curl ip.alo.is … $ curl ip.alo.is/json | jq .org …
iOS · App Store
NetTracer
A traceroute in your pocket: follow your network packets hop by hop, right from your iPhone. No account, no tracking, no ads.
05ContactLet’s talk about your infrastructure.
Tell me in a few lines about the system you are building or operating, and what is getting in the way. I reply within two working days.
Emailhello@alo.isOffice Freiburg im Breisgau, Germany