The modern QA stack
The tools available to QA engineers in 2025 are dramatically better than what existed five years ago — but also more fragmented. Here is what's actually being used and why.
Test management
Legacy: Excel spreadsheets, TestRail, Zephyr
Modern: softAssert, Linear, Notion
The shift is toward tools that integrate with development workflows rather than sitting in a separate silo. When your test cases live in the same system as your tickets and bugs, context is preserved and hand-offs are faster.
Bug tracking
Legacy: Email chains, Jira with 50 custom fields nobody fills in
Modern: Linear, GitHub Issues, Jira with AI-assisted field population
The best bug reporters I know write great bug reports because they have a system, not because they're naturally gifted writers. Templates + AI severity suggestions = consistent reports without the cognitive overhead.
Test execution
Automated:
- Playwright — the clear leader for end-to-end tests in 2025
- Vitest / Jest for unit and integration
- k6 for load testing
Manual:
- softAssert for structured suite execution with pass/fail tracking
- Screen recording tools for capturing reproduction steps
- BrowserStack for cross-browser coverage
AI tools actually being used
This is where things have changed the most in the last 12 months.
For test case generation: softAssert, GitHub Copilot (for automated test code)
For bug description: AI severity suggestions (softAssert), Grammarly for clarity
For root cause analysis: Claude or GPT-4 for analyzing stack traces and logs
For test data: AI-generated synthetic data that respects constraints
The stack gap: between manual and automated
The biggest productivity gap in QA today is between manual testing and automated testing. Manual is flexible but slow. Automated is fast but brittle and expensive to maintain.
AI is starting to fill this gap:
- Generate Playwright test skeletons from manual test cases (softAssert export)
- Identify which test cases are worth automating based on run frequency and failure rate
- Suggest which areas of the codebase lack test coverage
The QA engineers who are most productive in 2025 aren't fully manual or fully automated — they're strategic about which tests live in each layer.
What to cut
The tools that used to be standard but are worth reconsidering:
TestRail: Powerful but expensive for what you get. Most teams use 20% of the features.
Selenium: Playwright is strictly better in almost every scenario.
Dedicated defect management systems: When bugs live in Jira and test cases live in TestRail and the connection between them is manual, things fall through the cracks. Consolidate where possible.
The 80/20 stack
If I had to recommend a starting point for a QA engineer joining a new team in 2025:
- 1softAssert for test case generation, bug reporting, and suite management
- 2Linear or Jira for issue tracking (whichever the dev team uses)
- 3Playwright for automated regression tests on the top 10 user journeys
- 4BrowserStack for cross-browser and cross-device coverage
- 5Loom or Kap for recording reproduction steps
Everything else is optional. Start simple, add complexity only when you have a specific problem that demands it.