Free 20-minute call
We talk about where you are and whether I'm the right person to help. No commitment.
Compressing Complexity
I help CTOs of Swiss and EU scaleups get their platform ready for growth. Your Kubernetes, cloud and GitOps setup should scale without costs or complexity getting out of hand.
Services
Self-healing Systems
- Kubernetes, Helm, Kustomize, Flux
- Continuous deployment of applications and infrastructure
- Reliability, whether on bare metal or serverless
Developer Enablement
- Reassessing software and system architecture
- Creating and running bespoke workshops
- Facilitating developer involvement in infrastructure
Reliable Production
- Observability: instrumentation, metrics, logs, analysis
- Highly available and multi-master databases
- Blameless postmortems and root cause analysis (RCA)
When to reach out
Get in touch if:
- Your strong team has hit an unusual edge case
- Your cloud bill is growing faster than revenue
- You've just raised and need your platform to keep up
Working together
Say your team has to move an unusual setup off a deprecated Kubernetes API before the provider's support runs out. While we work through it, I notice your node pools still run a six-year-old machine family. You came for the migration and leave with a smaller cloud bill, too.
Most engagements start the same way:
Discovery
A one-hour session where you walk me through your setup, then a short written findings report. Book the Assessment within 30 days and the fee is credited.
Assessment
Fixed or results-based price. The report is usually ready about a week after you give me read-only access.
I start with a light scan across cloud cost, reliability, security basics, infrastructure as code, CI/CD, and team operations. Then I go deep on the one or two areas that matter most to you, chosen during Discovery.
You get a written report, a prioritised roadmap and a one-hour readout with your team, plus scoped and priced follow-up ideas to do in-house or with me.
Optional ongoing support
A set number of days per month, month to month with 30 days' notice. I work alongside your engineers on the hard parts, including introducing GitOps until your team runs it on its own. Your team stays in control throughout. Prefer no standing arrangement? Bring me in for a one-off, subject to availability.
Paying for results. Where results can be measured, such as cloud savings, you can tie part of the fee to the savings you actually get.
Read further
Approach
I try to understand both where you are now and where you need to get to. Sometimes my advice is as simple as finding a coding error or a technical misunderstanding. Other times I can suggest a less obvious alternative, or even a different way of framing the question. If I spot an unrelated issue I can help with quickly, I'll show you what's possible straight away, without waiting to rewrite the scope.
Helpful from the first session
I try to leave you with something you can act on every time we talk, even if you don't book anything else. If I can point you to the actual solution right away, even better: that's a success.
Your team stays in charge
I'd rather unblock your engineers and work through the problem with them than take over. Write access stays with them by default.
Short path to production
If you'd like your clusters rebuilt from scratch, Terraform or OpenTofu bootstraps the cloud resources and initialises Flux. Flux makes sure everything merged in git is also present and running in the cluster. That's one tool per layer and one place to look when something breaks. Of course, that's not the only way. I use it as my default for new builds, and I'm just as happy to work with what you already run.
Observability
Some problems only show up in real data. I'll help you instrument your applications and systems, with metrics in Prometheus and logs in Loki or whatever you already use. If you already collect that data, I can help you get more out of it, for example with Grafana dashboards that put metrics and logs side by side.
Simple systems, simple terms
The goal is a platform your team can run on its own, without big-firm overhead or lock-in.
About me
Hi, I'm Kris Ilowiecki. I've been programming since I was 11, and got my first job writing C for astrophysics in 2008. A couple of years later I took on university Unix sysadmin duties, joining what we called SysDev and applying an engineering approach that would later be known as SRE.
I moved to Switzerland for a fintech job in 2013, ran virtual power plants on AWS by 2015, and made the jump to Kubernetes in 2017. By 2018, as a two-person team, we ran over 6'000 containers concurrently.
I've built clusters and other production infrastructure across the three major hyperscalers, while introducing SRE principles to a number of startups and scaleups.
For the three years before I founded Scaling Horizon, I worked as a Senior Cloud Architect at DoiT International. I advised CTOs and top technical teams of cloud-native customers across industries and ran Kubernetes workshops for hundreds of engineers.
I seek out difficult challenges and tackle them head-on. Every situation is different, so it helps that I understand the whole stack, from shift registers through machine code, syscalls and protocols up to the orchestrator. I also draw on what I learned during my physics degree (MPhys, Oxon).
If your team is stuck on something unusual, I'm happy to take a look. Whether your infrastructure spend is in the tens of millions or you're just getting started, we can learn from each other.
Contact
The easiest way to start is a free 20-minute call. Tell me a bit about your setup and what's slowing you down, and I'll get back to you.
- mailto:
- kris@scalinghorizon.ch
- linkedin://
- linkedin.com/
in/ kilowiecki