Skip to content
Jan Podzimek

QA Engineer · Alza.cz

Jan Podzimek

I test web applications, build API test automation in C#, and design AI agents that make QA faster, from test analysis to a tool that runs the release process for the whole team.

  • 4+ years in QA
  • ISTQB® CTFL 4.0
  • English C1
Download CVGet in touchLinkedInGitHub
Portrait of Jan Podzimek

About me

I’m a QA engineer at Alza.cz, part of a four-person QA team covering two web applications. Over four years I’ve tested front end, back end, integrations and end-to-end flows, and I adapt to whatever the work needs.

My testing is context-driven. I put critical thinking, exploration and real-time analysis ahead of rigid scripts, and I choose the technique the situation calls for instead of following predefined test cases to the letter.

I don’t like splitting testers into “manual” and “automation”. Good testing combines thinking, exploring and the smart use of tools. Lately, a big part of my tools has been AI agents I designed and built myself.

Experience

QA Engineer · Alza.cz a.s.

2022 – present

Member of a 4-person QA team covering two web applications.

  • AI-assisted testing: designed and set up custom GitHub Copilot agents (test analyst, test designer, documentation writer) that generate test analyses, test scenarios and application documentation.

  • API test framework: built the team’s first API test framework from scratch in C# and NUnit, running in the Azure DevOps CI pipeline.

  • Release automation: designed an internal tool and built it through agentic coding. It has largely replaced the manual release process.

  • Everyday QA tooling: on-demand deployment of any feature branch to a chosen test environment, plus utilities for daily testing, used by the whole QA team.

  • Test analysis & execution: evaluate new features, run functional and regression testing, and verify data and integrations using MS SQL, Kibana and RabbitMQ.

  • Release management: coordinate releases and monitor deployments and production stability (Argo CD, Kibana).

  • Mentoring: onboarding buddy for new team members.

Selected work

Three things I built at work. Internal details are left out; the problem, approach and outcome are real.

case-01

Release automation tool

An internal tool, built through agentic coding, that took over most of the manual release process.

Stack

  • Agentic coding (GitHub Copilot)
  • Azure DevOps
  • Argo CD
  • Microsoft Teams
Problem
Every release meant the same manual routine: creating release-candidate branches in several repositories, opening and merging pull requests, deploying to the dev environment and telling the team. It was slow, repetitive and easy to get wrong.
Approach
  • Designed the tool around the team's real release checklist.
  • Built it through agentic coding with GitHub Copilot. I defined the behaviour, then reviewed and tested every change.
  • It creates RC branches across repositories and opens and merges pull requests with descriptions.
  • It deploys the release candidate to dev for regression testing and posts a notification to the Teams release channel.
Result
  • Largely replaced the manual release process.
  • Extended with on-demand deployment of any feature branch to a chosen test environment, used by the whole QA team.

case-02

The team's first API test framework

Built from scratch in C# and NUnit and wired into the Azure DevOps CI pipeline.

Stack

  • C#
  • .NET
  • NUnit
  • HttpClient
  • Azure DevOps Pipelines
Problem
Before each release the APIs were checked by hand. That cost time on every release, and there was no automated safety net to catch regressions early.
Approach
  • Designed the framework from scratch in C# and NUnit, using HttpClient.
  • Added health-check coverage of every endpoint as a fast first line of defence.
  • Built functional, authorization and validation suites on top.
  • Runs automatically in the Azure DevOps CI pipeline.
Result
  • Cut manual testing time before each release.
  • Gave the team a shared base that new API tests plug into.

case-03

AI agents for test analysis

Custom GitHub Copilot agents that help the team with test analysis, test design and documentation.

Stack

  • GitHub Copilot
  • Custom agents
  • Prompt engineering
Problem
Test analyses, test scenarios and application documentation take a lot of time, and their quality depends on who writes them and how much time they have.
Approach
  • Designed three specialised agents, a test analyst, a test designer and a documentation writer.
  • Gave each agent its own instructions, scope and output format.
  • Every output is reviewed by a tester before it is used.
Result
  • Faster first drafts of test analyses, test scenarios and documentation.
  • A more consistent structure across the team's analyses.

How I test

Before I design a single test case, I analyse the feature. This is the checklist I use, adapted from context-driven testing.

  1. 01

    Purpose & intent

    What problem does the feature solve, for whom, and what would make it useless even if it technically works?

  2. 02

    Tracking & monitoring

    Entry and exit points, CRUD history and error logs, so I can see what actually happened, not just what the UI shows.

  3. 03

    Behaviour analysis

    Incorrect setups, invalid data and unintended use. How does it fail, and does it fail safely?

  4. 04

    Dependencies & connections

    Which services, queues and data does it rely on, and what breaks when one of them is slow or down?

  5. 05

    Scenarios

    Realistic, complex user journeys instead of isolated clicks.

  6. 06

    Integration

    How it behaves as part of other functionality, and whether its parts work on their own.

Try my QA tools

Small utilities I use when testing, rebuilt for this site. Each one is covered by unit and UI tests.

/tools/column-delimiter

Column delimiter

Paste one record per line and get them joined into a delimited list, e.g. for a SQL IN clause or a Jira JQL filter.

Open tool →

/tools/char-counter

Character counter

Count characters, words, lines and bytes. Handles diacritics and emoji correctly, which is handy for testing field length limits.

Open tool →

All tools →

Skills & tools

Testing
Test analysis & design, functional, regression and API testing
Automation
C#, .NET, NUnit, HttpClient, Postman
AI
GitHub Copilot, Claude Code, agentic coding, custom agents for test analysis, test design and documentation
DevOps
Azure DevOps (repos, pull requests, pipelines), Git, Argo CD
Data & monitoring
MS SQL Server, Kibana, RabbitMQ
Collaboration
Jira, Confluence, Excalidraw (mind maps)

Certifications & education

Certifications

Education & training

  • Object-Oriented Programming in C#

    VŠB – Technical University of Ostrava · 05/2022 – 11/2022

    6-month course: C# and OOP, MS SQL Server, Git, HTML5/CSS3, test automation basics (MSTest, NUnit, Selenium).

  • Manual Software Testing

    Software Development Academy · 12/2021 – 04/2022

    Testing methodology per the ISTQB syllabus, test scenarios, test cases and bug reports.

Outside of work

Puzzles

Puzzles train the same skills as testing: patience, creativity and looking at a problem from a different angle.

Sport

Running, hiking, badminton, obstacle races, the gym. Anything that keeps me moving.

  • Resting on a mountain ridge after a hike
  • Hanging from an obstacle at a Gladiator Race
  • Playing guitar by a campfire

Let’s talk

I’m open to new opportunities and to conversations about testing, automation and AI in QA.

Send an emailLinkedInGitHub

podzimek.jan@seznam.cz