Workshops: Agentic Coding
Agentic Coding for QA & Test Automation: Workshop
AI agents can derive test cases, write end-to-end tests and analyse flaky tests, if QA teams guide them well. This workshop is for testers and test automation engineers who want to use agents to raise test coverage and stabilise pipelines.
Who is this workshop for?
- QA engineers and software testers
- Test automation engineers
- QA leads and test managers
- Development teams responsible for test quality
- Manual testers moving into automation
What you will take away
- Derive structured test cases from requirements and user stories with agents, including boundaries and negative cases.
- Have agents write Playwright or Cypress E2E tests using stable selectors and page objects.
- Analyse flaky tests and pipeline failures systematically with agent support.
- Generate realistic, privacy-compliant test data.
- Judge AI-generated tests on what they actually verify, not just on coverage.
Agenda
Module 1
Agents in the testing process
- Where agents help QA: test design, automation, analysis
- Claude Code, Cursor, GitHub Copilot and OpenAI Codex for test code
- Project rules for test conventions
Module 2
Test design with AI
- Exercise: derive test cases from user stories
- Equivalence classes, boundaries and negative tests
- Gherkin scenarios and acceptance criteria
Module 3
E2E automation
- Writing and running Playwright tests with agents
- Playwright MCP for agent-driven browser control
- Page objects, fixtures and stable selectors
Module 4
Stability and analysis
- Reproducing flaky tests and finding root causes
- Having CI logs analysed
- Mutation testing to assess test quality
Module 5
Test data and governance
- Synthetic test data instead of production data
- Reviewing AI-generated tests
- The QA role in agentic development
The new role of QA
As agents write more code, the need for reliable verification grows. QA becomes the function that defines what correct means and gives agents the tools to check it themselves.
At the same time, agents easily produce tests that cover a lot but assert little. We train the critical eye that separates good tests from worthless ones.
Tools in the workshop
We work with your test frameworks and add agent tooling on top.
- Playwright, Cypress, Selenium
- JUnit, pytest, Jest, Vitest
- Playwright MCP and browser automation
- CI systems such as GitHub Actions, GitLab CI or Jenkins
Format
Delivered by neooutput AI Code Engineers with QA experience. Remote or on-site, in English or German, using your application as the practice target.
Frequently asked questions
Do participants need coding skills?
Basic knowledge of a programming language helps. Manual testers benefit especially from the test design part and learn how agents can write automation code for them.
Will AI replace our testers?
No. AI takes over routine work in writing and maintaining tests. Domain understanding, risk assessment and exploratory testing remain human tasks.
Can we use our staging environment?
Yes. We recommend a test or staging environment without real customer data and agree on access with your IT beforehand.
Workshop
Get in touch
Want agents to raise your test coverage without diluting what tests prove?
Tell us a little about your team – we will get back to you within 24 hours with a proposal for content, format and dates.
Prefer to talk?
More workshops
- Workshops: Agentic CodingAgentic Coding for Data Science & Analytics: Workshop
- Workshops: Agentic CodingAgentic Coding for DevOps & Infrastructure as Code: Workshop
- Workshops: Agentic CodingBuilding MCP Servers: Hands-On Workshop for Dev Teams
- Deploy & App StoresHow to Publish a Bolt.new App on the Apple App Store
