Kubernetes consulting and managed Kubernetes
Kubernetes consulting from a team that runs production clusters.
We fix, upgrade and migrate your clusters, or take them over and run them, in your cloud account or on your own hardware.
Kubernetes consulting is outside help to design, build, fix or run Kubernetes clusters. A Kubernetes consultant reviews your clusters, how you deploy to them and what they cost, fixes what is fragile or expensive, and then hands the work back to your team or keeps running it.
We do both. As consultants, we fix and build while your team runs the clusters day to day. As a managed Kubernetes provider, we run them for you. Either way, you meet the lead engineer before you sign. The lead is your contact and the team behind them does the work, so your clusters don't depend on one person.
Why teams call us about Kubernetes
Upgrades nobody wants to touch
The cluster is several versions behind and every upgrade feels like a bet with production. On EKS and GKE, a version past standard support also raises the cluster fee from $0.10 to $0.60 an hour.
One person who understands the cluster
Pods restart, nodes fill up, and the engineer who set it all up is on vacation or has left.
A bill that keeps growing
Requests sized just in case, idle nodes and load balancers nobody remembers creating.
What our Kubernetes consulting covers
Cluster design and setup
Production clusters on EKS, GKE, AKS or your own hardware, with high availability, storage and networking designed for your workload.
Upgrades
Version upgrades planned and tested before they reach production, so the cluster stays on a supported release.
Migration
Moving applications onto Kubernetes, in the cloud or in your own data center, in stages that keep production running.
CI/CD and GitOps
Deploy pipelines with Argo CD and Helm, or in GoCD, GitHub Actions and GitLab CI, so every change to the cluster goes through Git and can be rolled back.
Security
RBAC, network policies, secrets and image scanning. We harden the clusters; for a SOC 2 or ISO 27001 audit you still need an auditor.
Cost optimization
Right-sized requests, autoscaling and better node choices. Our cost optimization checklist shows where the money usually goes.
Monitoring and on-call
Prometheus and Grafana with alerts on what your users notice, and on-call by agreement.
Handover
Documentation, runbooks and walkthroughs, so your team can run what we built without us.
Managed Kubernetes: we run your clusters
Managed Kubernetes from a cloud, such as EKS, GKE or AKS, runs the control plane, and on some plans the nodes. Everything else is still yours, and that is the part we take over, in your cloud or on your own hardware.
Upgrades
Kubernetes versions, add-ons and node images kept on supported releases.
Ingress, certificates and storage
Ingress controllers, TLS certificates and persistent volumes that keep working as the cluster changes.
Monitoring and incidents
Alerts on what your users notice and someone to answer them, around the clock if you need it.
Security patches
Nodes, images and add-ons patched, and access reviewed as people join and leave.
The bill
Requests, autoscaling and idle resources kept in check, so costs don't creep back up.
On-prem control plane
On your own hardware no cloud runs the control plane, so we run it too: etcd backups, certificates and upgrades.
Ways to work together
Emergency help
A cluster is down or an upgrade went wrong. We step in, get it running again and tell you what happened.
Assessment
A fixed price. We review your clusters, deployments, security and costs, and give you a written list of problems ordered by what each one costs you.
Project
A fixed or capped price for a defined piece of work, such as an upgrade, a migration or a new cluster, done by the team that scoped it.
Fixed price for a result
We agree on the result and the deadline, and you pay a fixed price for the result, however many hours it takes us.
Ongoing support
We run your clusters for a monthly fee, month to month or on a yearly contract. By agreement it includes on-call, up to 24-hour coverage under an SLA.
Not sure which fits? Send us a few lines.
How a Kubernetes engagement starts
- 1
Intro call
Tell us where the clusters run and what hurts. Before anything is signed, you meet the lead engineer who would be your contact.
- 2
Review
With read-only access, we study the clusters, manifests, pipelines and bills without changing anything, and write down each problem and what fixing it involves. If Kubernetes is more than you need, we say so.
- 3
Fixes
The team that did the review makes the changes, through your repositories and your code review.
- 4
Handover or we run it
We document what we built and walk your team through it, or we keep running the clusters for you.
Kubernetes consulting vs. managed Kubernetes, freelancers and an in-house team
| Option | What you get | Good fit when | Trade-off |
|---|---|---|---|
| Cloud managed Kubernetes (EKS, GKE, AKS) | The cloud runs the control plane, and on some plans the nodes | You want less to operate and are settled on one cloud | Upgrades, add-ons, security and the bill are still your team's job |
| Kubernetes consulting | Engineers review, fix or build, then hand it back | Your team will run the clusters day to day | Your team needs the time and skills to keep it running |
| Managed Kubernetes service provider | A team runs your clusters: upgrades, monitoring, incidents and costs | You want the clusters run without hiring a platform team | You depend on the provider, so agree on access and handover up front |
| Freelance Kubernetes consultant | One expert, usually billed by the hour | A defined task you can review yourself | When they are away the work waits, and what they know leaves with them |
| In-house platform team | Your own engineers | Kubernetes is core to the business and you can hire for it | Senior hires take months, and a small team is a single point of failure |
Many teams combine them: a cloud runs the control plane, and a consultant or a provider covers the rest.
We have run production systems since 2009 and have managed more than 1,350 production servers. We moved a long-term client's big-data platform onto on-prem Kubernetes, and we still run those clusters: version upgrades, and moving their databases in.
Tools we work with
Frequently asked questions
- How much does Kubernetes consulting cost?
- It depends on how many clusters you run, what is broken and whether you need on-call. We work at a fixed price for an assessment, a fixed or capped price for a project, a fixed price for an agreed result, or a monthly fee for ongoing support. We quote after the intro call.
- What is the difference between Kubernetes consulting and managed Kubernetes?
- Consulting is a project or advice: experts fix or build, and your team runs the clusters afterwards. Managed Kubernetes means someone else runs them. From a cloud such as EKS, GKE or AKS, that covers the control plane. From a service provider like us, it covers the rest too: upgrades, monitoring, incidents and costs.
- How long does a Kubernetes project take?
- It depends on the clusters and what is broken, so we don't quote a timeline before the review. With a fixed price for a result, the deadline is agreed up front.
- Do you offer 24/7 Kubernetes support?
- If you need it, yes. We agree on the SLA up front: how fast we respond to a problem, and how long systems can be down when the problem is in our scope.
- Can you run Kubernetes on our own hardware?
- Yes. We run on-prem and bare-metal clusters as well as cloud ones. Without a cloud provider, nobody runs the control plane for you, so we cover it too: etcd backups, certificates, upgrades and storage. Our post on bare metal vs. private cloud covers the trade-offs.
- Who owns the clusters and the code?
- You do. The code and configuration we write belong to you, the clusters run in your cloud or on your hardware, and you can take the work in-house.
- What is the difference between Kubernetes consulting and DevOps consulting?
- Kubernetes consulting covers the clusters: design, upgrades, security, cost and day-to-day operations. DevOps consulting covers the whole path from code to production, and Kubernetes is one part of it.
- Is Kubernetes still relevant in 2026?
- Yes. In the CNCF's 2025 survey, published in January 2026, 82% of container users ran Kubernetes in production, up from 66% in 2023. Among organizations hosting generative AI models, 66% used it for some or all of their inference.
- Why are companies moving away from Kubernetes?
- Often because it was more than they needed. Kubernetes adds moving parts, and a small team running a few services can spend more effort on it than it gets back. Some workloads fit badly: Gitpod, now Ona, said in 2024 that it was leaving Kubernetes for its development environments, which behave nothing like the application workloads Kubernetes was built for. If a simpler setup fits you, we will say so.
- Which provider offers the cheapest managed Kubernetes?
- DigitalOcean and Akamai (Linode) don't charge for a standard control plane (high availability costs extra), and Azure AKS has a free tier without a financially backed SLA. Amazon EKS and Google GKE charge $0.10 per cluster per hour, about $73 a month, though GKE's free tier covers one zonal or Autopilot cluster per billing account (prices as of October 2026). The control plane is the small part of the bill: nodes, load balancers, storage and traffic make up most of it.
- Are there Kubernetes certifications?
- Yes. The CNCF and the Linux Foundation run the main ones: CKA for administrators, CKAD for application developers and CKS for security, plus the entry-level KCNA and KCSA.
From our blog
View all posts →
Kubernetes Cost Optimization: A Practical Checklist
Kubernetes cost optimization, step by step: cost monitoring, right-sized resource requests, autoscaling, Spot, idle resources and the tools for each.
Self-hosted K8s – Bare metal vs private cloud
Kubernetes began as an internal Google project, an orchestration system for “containers” — OS-level virtualization instances of larger applications, usually...
You might also need
Tell us about your clusters
A few lines are enough. We will tell you whether we can help and what we would look at first. If it fits, we start with read-only access and a written list of what to fix first.
- Where the clusters run: which cloud, on-prem or both
- What hurts: upgrades, outages, costs or one person who knows it all
- Whether you want it fixed and handed back, or run for you