I build test structure, test automation and AI-driven test generation into your development. In your environment, with your team. Not as a service you rebook every quarter.
Most teams that call me don’t have a bad product. They just never decided how to test it. In practice, it looks like this:
Whatever stands out gets tested. Whatever doesn’t, doesn’t.
First the feature, then the deadline, then testing, if there’s time left.
Hours of manual clicking, and the bug still slips through.
Your team knows how to build. Nobody ever taught them how to test.
The result is always the same: at some point, your customer finds the bug before you do. A bug that only surfaces in production costs many times what you would have invested in an early test. And it costs trust.
When a ticket isn’t finished in time because the bug only shows up in testing, or even later, the sprint slips. That’s €13,750 of development capacity going into a delay instead of into your product. Four times a year: €55,000.
Full cost incl. employer contributions, 220 working days per year.
You’re not paying too little for QA. You’re paying more every year for the same result. There’s a mechanical reason for that.
Every new feature enlarges the surface that has to be checked before each release. No feature ever shrinks it. Your testing effort grows with the age of your product, not with your revenue.
Scale QA by adding people and you get a curve that never flattens. The metric that makes this visible is Cost per secured release. Almost nobody measures it.
Automation solved test execution. The last manual bottleneck is writing and maintaining test cases. That’s exactly where the curve breaks, and exactly where I come in.
QA isn’t an extra cost line in your budget. It’s an investment that stops you from paying for the same work twice.
Automation without structure just automates the chaos. That’s why I work in a fixed order.
First we clarify what needs testing, then build a strategy that fits your team.
Before anything gets automated, we clarify what actually needs testing. I review your requirements, your existing tests and your development process, and build a test strategy that fits your team, not a textbook.
Tests that run on every commit instead of costing someone time on every release.
Tests that run on every commit instead of costing someone time on every release. I build the automation and wire it into your pipeline so a red result is a clear signal, not a reason to switch the test off.
AI writes more code than ever, but hardly anyone checks systematically whether it does what it should. I advise you on using AI in your quality assurance the right way.
This is where the difference lies, in both directions. I advise on AI in testing and put it to work: your requirements become test cases and automation code. And I test what AI builds, from generated code to AI products. A test case an AI writes in seconds is only worth something once a human has checked that it tests the right thing.
test('checkout')
expect(total).toBe(…)
The manual step in the middle disappears. And the generation runs in your environment, not mine.
Three situations that come up again and again in different forms. Client names are withheld for confidentiality.
ISTQB certified · Project experience in large enterprises and the public sector, including Deutsche Bahn, Bilfinger and Paychex
Test service providers use AI too these days, to speed up their own work. You get a result. I build AI into your process. You get a capability.
In concrete terms: a test generation system that creates test cases from your requirements and runnable automation code from those. It runs in your environment, not mine. When I leave, your team can do what I do.
You work with me directly. No junior taking over after the sale, no middle layer, no handoffs. That’s also why I only work with a few clients at a time.
| Your alternative | What stays afterwards |
|---|---|
| Hiring another QA engineer | one more head |
| A tester from a consultancy | nothing |
| Working with me | the structure and the AI system |
Your alternative is rarely “no QA at all”. It’s hiring another person or buying someone in. Both give you capacity, and a new role takes time to build up. A consultant focused on AI testing is up to speed on day one. Here you’re investing in a capability that stays when the contract ends.
If you only want an outside opinion to tick a box, save your money.
QA structure pays off over months, not two weeks.
I need access to your code, your processes and your open issues. Without that, it doesn’t work.
You describe the situation, I ask questions. By the end, you know whether I’m the right fit. Free, no obligation.
We go into your process, your tests and your pipeline in detail. This is where the problem turns into a scope.
You get a proposal with a clear scope and a clear price.
From the first call to a signed proposal: possible in under two weeks.
I discuss pricing in the call, not on the website. Not as a tactic, but because the right investment depends on scope: a QA structure for a startup with four developers is something different from a regression strategy in a large enterprise. A number without context would be a guess.
Thirty minutes, free. You tell me where things get stuck, and I’ll tell you honestly whether and how I can help. If it’s not a fit, I’ll tell you in the call.
Book an intro callOr send me an email: info@kdessel.com