Infrastructure that doesn’t wake you up
at 3
AM.
When systems misbehave and alerts pile up, we bring calm, clarity, and fixes that stick. From legacy migrations to modern automation — we make infrastructure reliable, predictable, and quietly boring.
root@sayupdate:~# ./incident.sh
Detecting issue... [OK]
Checking DNS (of course)... [OK]
Applying fix... [OK]
Production stable.
root@sayupdate:~# _
Tools we trust
Linux
Docker
Kubernetes
Terraform
AWS
Ansible
Bash
GitLab
Prometheus
GrafanaServices
Pick one area — or hand us the mess. We’ll triage, fix, and leave things better than we found them.
Bare Metal & SysOps
Real servers, real constraints. BIOS, firmware, RAID, networking, and OS installs — the unglamorous foundation that everything else depends on.
Hybrid Cloud & Virtualization
Cloud where it helps, on-prem where it makes sense. We design and operate hybrid setups using Proxmox/KVM and public cloud without forcing everything into one model.
CI/CD & Automation
Predictable pipelines you can actually debug. Terraform for infrastructure, Ansible for configuration, and CI/CD workflows designed to be owned — not feared — by your team.
Incident Response & Troubleshooting
When production breaks, we go deep: logs, metrics, networking, kernel quirks. Fast isolation, clear fixes, and post-mortems focused on improvement — not blame.
Cost & Performance Optimization
Lower bills without fragile shortcuts. Right-sizing, architecture cleanups, and finding the “why is this running?” resources that quietly burn money.
Backups & Disaster Recovery
Because "it worked yesterday" isn't a recovery plan. We implement automated, off-site backups and — more importantly — test the restores.
How We Work
No rituals. No heroics. Just a clear process that works when things are calm — and when they aren’t.
1. Understand
“First, we read the logs.”
Before proposing fixes or tools, we learn how your system actually works today — architecture, constraints, traffic patterns, metrics, and alerts. No assumptions. No “just rewrite it.”
2. Fix What Hurts
“DNS jokes optional. Root causes mandatory.”
We focus on the biggest sources of pain first: instability, slow delivery, scaling limits, or runaway costs. Quick wins where possible, solid foundations where needed.
AI (Used Responsibly)
“AI is fast. Context is slow. We provide the context.”
We use AI to accelerate analysis, spot patterns, assist with documentation, and support automation. But good results don’t come from asking better questions — they come from knowing what to ask, when to ask, and what to ignore.
AI speeds things up. Experience keeps it from doing something creative in production.
*No models were deployed directly to production without supervision.
3. Automate
“If it’s manual twice, it’s a bug.”
Infrastructure as code, repeatable deployments, and automation that removes humans from repetitive, error-prone tasks — especially at night.
4. Leave It Better
“No black boxes. No magic scripts.”
Clear documentation, explained decisions, and systems your team can confidently own. You shouldn’t need us forever — but you’ll know when to call us back.
Michał Michulec
Lead DevOps Engineer & FounderSAYUPDATE is a boutique DevOps consultancy focused on the work that’s often postponed until it becomes urgent: legacy migrations, fragile deployments, and systems that demand constant attention.
We combine old-school SysOps discipline with modern DevOps practices to build infrastructure that is understandable, maintainable, and resilient.
"Boring infrastructure is a feature, not a failure."
Let’s make your infrastructure boring
Tell me what you’re running, what hurts, and what “good” looks like. I’ll help you get there — whether that’s a quick fix or a longer-term plan.
[email protected]