I test software.
I test AI systems.
I test with AI.
I'm a QA Engineer who goes beyond classic testing — I check the quality and safety of AI-based systems, and I build and use AI tools to test faster and more effectively.
Three pillars — one competency: quality
Classic QA and AI testing are one field today. I combine both to give a complete picture of product quality.
QA Testing
Full quality assurance cycle — from test planning to defect reporting.
- Manual functional & exploratory testing
- Test automation (web, API)
- Regression, edge cases, test documentation
AI Systems Testing
Verifying the quality, consistency and safety of solutions built on language models.
- Testing LLMs, chatbots and AI agents
- Evaluating hallucinations, consistency and prompt regressions
- Testing AI integrations (APIs, tools, workflows)
AI in the Testing Process
I use AI tools to speed up and strengthen the classic QA process.
- Generating and maintaining test cases with AI
- Building custom AI-based tools and agents
- Hands-on experience with Claude, Cursor and similar tools

From software quality to AI quality
Day to day I work as a QA Engineer and QA Team Lead, but I don't stop at classic testing. I'm interested in the intersection of QA and artificial intelligence — both testing systems built on AI, and using AI as a tester's own tool.
This isn't theoretical knowledge — on my own projects I build and maintain AI-agent-based tools (including an autonomous task management system with multiple AI reviewers, and an automatic video overlay generator), so I know first-hand how such systems fail and how to actually test them.
Testing news, research & realities
AI is reshaping software testing faster than most teams can adapt. Here's what the data actually says — and why it matters for how I test.
The market is moving fast
AI-powered testing tooling is one of the fastest-growing corners of software quality — but most organizations are still piloting, not scaling. Widespread experimentation, uneven results: that gap is exactly where careful, hands-on QA still matters most.
Hallucination rates aren't one number
Depending on the benchmark and task, published hallucination rates for language models range from single digits to over 90%. Headline percentages mean little without knowing what was actually measured — which is why I test against the specific failure modes that matter for a given product, not a generic benchmark.
More AI-written code means more testing demand
A growing share of production code is now AI-generated or AI-assisted. Teams report that this increases — not decreases — the amount of testing needed, since AI-written code introduces its own class of subtle, non-obvious bugs.
Testing services — order what you need
End-to-end test coverage for your product, quoted individually — from a single test pass to ongoing QA support.
Types of testing
- Manual functional testing
- Test automation
- Regression testing
- Exploratory testing
- Performance testing
- Security testing (baseline)
- Usability / UX testing
- Accessibility testing (WCAG)
- API testing
- AI / LLM testing
- Test case management (TestRail, Testify, etc.)
Platforms & devices
- Web (desktop & mobile browsers)
- iOS & Android apps
- Desktop applications
- Backend / API services
- Real devices & emulators
- Cross-browser & cross-OS coverage
Project types
- Web applications
- Mobile apps
- Marketing & landing pages
- E-commerce platforms
- Internal / B2B systems
- AI-powered products & chatbots
Don't see your case listed? Every engagement is scoped and quoted individually.
Let's talk about your product's quality
I'm open to projects, consulting and ongoing collaboration — write or call.