Jarosław Sitarczuk
Book a call

DevOps for startups that don’t have one.

I set up, fix, and run infrastructure for small teams: CI/CD, cloud, deploys, monitoring, backups. Fixed-price packages or a monthly retainer. You talk to me, not an account manager.

Book a 30-min callNo pitch. We look at your setup and I tell you what I’d do.

Azure · Kubernetes · GitLab CI · GitHub Actions · Terraform · Docker · Linux

Sound familiar?

  • Deploys are a ritual. One person knows how, and everyone hopes they’re not on holiday.
  • The cloud bill grows every month and nobody knows exactly why.
  • “We’ll add monitoring later.” It’s been later for a year.
  • Someone set up the infra in a weekend two years ago. It works. None of it is in code, so nobody touches it.
  • You can’t justify a full-time DevOps hire, but the problems don’t care.

This is the work I do. Usually in a week or two, sometimes on a monthly basis.

Packages

Fixed scope, fixed price, no hourly surprises. Prices are net (excl. VAT).

Cloud cost audit

900 EUR / 3 000 PLN

For: teams spending 2 000 EUR/month or more on Azure (or AWS/GCP).

You get
  • Full review of your cloud spend, resource by resource
  • Quick wins implemented during the audit (rightsizing, unused resources, reservations)
  • Short report: what to change, what it saves, what it costs
Time:
1 week
Guarantee:
If the audit doesn’t identify at least the audit fee in annual savings, you don’t pay.

CI/CD setup or rescue

from 1 500 EUR / 5 000 PLN per project

For: teams deploying by hand, or with a pipeline nobody trusts.

You get
  • Pipeline on GitLab CI or GitHub Actions: build, test, deploy
  • Separate environments (staging / production) with proper secrets handling
  • Rollback that actually works
  • Handover session and written docs
Time:
1–2 weeks

Production readiness review

1 200 EUR / 4 000 PLN

For: teams about to launch, or already live and not sleeping well.

You get
  • Review of infra, security basics, backups, monitoring, access, single points of failure
  • Check whether the setup is reproducible from code or lives in someone’s head
  • Prioritized list: fix now / fix this quarter / fine for now
  • 1-hour walkthrough with your team
Time:
1 week

Zero to production

from 2 500 EUR / 8 000 PLN

For: early startups with a working app and no real infra yet.

You get
  • Cloud setup sized for your stage (not the biggest thing, the right thing)
  • Everything as code (Terraform), in your repo, not mine
  • CI/CD, environments, deploys
  • Monitoring, alerting, backups
  • Written docs your next hire can read
Time:
2–3 weeks
Main offer

Fractional DevOps (retainer)

  • 1 800 EUR / 5 500 PLN per month for 20h
  • 3 200 EUR / 10 000 PLN per month for 40h

For: teams that need someone owning the infra, without a full-time hire.

Terms: monthly, cancel anytime with 30 days notice.

You get
  • A senior engineer who owns your infra, async, on your Slack
  • Ongoing improvements, fixes, upgrades, cost control
  • Infrastructure as code from day one, so anyone can take over
  • Monthly summary: what I did, what’s next, what’s at risk
  • Priority over package clients
Book a 30-min call

Not sure which one? Book the call, I’ll tell you. Sometimes the answer is “you don’t need me yet”, and I’ll say so.

How it works

  1. 01

    30-min call.

    You show me the setup, I ask questions. No sales pitch.

  2. 02

    Written scope within 48h.

    What I’ll do, what’s out, price, timeline.

  3. 03

    You get access, I start.

    I work async. Everything I change goes into code and docs, in your repos.

  4. 04

    Handover.

    Docs, a walkthrough with your team, and I’m available for questions after.

Recent work

Anonymized. Happy to talk details on a call.

Startup on Azure, built by clicking

Problem

Two years of infrastructure created by hand in the Azure portal. Nobody could say what existed, what it cost, or what would break if touched. Pipelines scattered, secrets in the wrong places.

What I did

Inventory of everything, then the whole setup rebuilt in Terraform. Unused and oversized resources cleaned up, access and network locked down, all pipelines moved to GitHub Actions.

Result

33% lower monthly Azure bill. Every resource in Terraform, every change goes through a pull request.

Product teams on bare VPS and on-prem

Problem

Small companies running production on a few servers, set up once and never revisited. No monitoring, ad-hoc deploys, backups nobody had tested.

What I did

Full setup from the OS up: hardening, firewalling, containerized deploys, monitoring and alerting, tested backups and restores, written runbooks.

Result

Six environments running without incidents for over a year, and a team that can deploy and restore without me.

Taking over a Kubernetes setup from another vendor

Problem

Client wanted to move away from the software house running their Kubernetes cluster and GitLab CI / GitOps setup. No documentation, unclear ownership, nobody on the client side willing to touch deploys.

What I did

Full analysis of the existing setup, migration and takeover plan with risks called out, handover executed step by step.

Result

Setup transferred with zero downtime. Documented end to end, and the client’s own team now ships to the cluster.

Who’s behind this

Portrait of Jarosław Sitarczuk

I’m Jarosław Sitarczuk, based in Lublin, Poland. Ten years in software: I’ve written frontend, Node and Go backends, and for the last few years I’ve been doing DevOps for multiple clients across Azure, Kubernetes, and GitLab/GitHub CI. I’ve also interviewed and hired backend developers, so I know what a dev team actually needs from infra: less magic, more “it just deploys”.

I work alone on purpose. You get one senior person, the same one every time, who answers on Slack and doesn’t bill you for meetings about meetings. Everything I touch ends up in code, in your repositories, so nothing stays only in my head.

Registered business in Poland, EU VAT invoices, contracts in English or Polish.

Need more than infra?

For feature work, frontend, or product ownership, I bring in people I’ve worked with for years. Same setup: no agency layer, you talk to the people doing the work.

MVP build

Product owner + frontend + Go backend + infra, from idea to running product. Quoted per project.

Ask about it

FAQ

Do I need Kubernetes?
Probably not yet. Most teams under 20 people are better off with something simpler. I’ll tell you honestly which is right for your stage and I set up both.
Which clouds?
Azure is where I’ve done the most work. AWS and GCP are fine for audits and CI/CD. I also work with plain VPS and on-prem setups.
Do you work hourly?
Only inside the retainer. Everything else is fixed price, so you know the cost before we start.
What if the package doesn’t fit?
Then I’ll write you a custom scope. The packages cover the common cases, not every case.
What happens when we stop working together?
You keep everything. Infrastructure is in Terraform in your repo, pipelines are in your repo, docs are written. Nothing depends on me being around.
Can you sign an NDA?
Yes.

Let’s look at your setup.

30 minutes, no pitch. You’ll leave with a clear idea of what I’d change, whether or not you hire me.

Book a 30-min call
Book a 30-min call