AI-Augmented Quality Engineering
Using AI not as a buzzword but as a force-multiplier for test strategy, automation, and team productivity.
Tools I Work With
Python
Primary language for test automation, scripting, and AI/LLM integrations — used across CI pipelines, data generation, and framework tooling.
JavaScript
Used in frontend automation with Playwright and Cypress, and for Node.js-based integration testing and test tooling scripts.
Java
Test automation with Selenium and JBehave/BDD frameworks — particularly in enterprise and banking-sector projects.
Playwright
End-to-end browser automation across Chromium, Firefox, and WebKit. Primary framework for modern web UI automation and CLI testing.
Cypress
Component and integration testing for JavaScript frontends — fast feedback loops, strong developer experience, and built-in network interception.
Selenium
Browser automation across complex enterprise stacks, including online banking and legacy web applications requiring cross-browser coverage.
GitLab
CI/CD pipeline configuration and management — automating test execution, coverage reporting, and deployment gates on every merge request.
Claude Code
AI pair-programmer for scaffolding test frameworks, generating test cases from specs, and refactoring automation code. Used daily as a core part of the development workflow.
GitHub Copilot
In-editor code completion for Python, JavaScript, and Java automation scripts — reducing boilerplate and speeding up pattern-based test writing.
OpenAI SDK
Integrating LLM-powered assertions and test data generation into automated pipelines — enabling intelligent, context-aware quality checks.
Prompt Engineering
Crafting precise prompts for test case generation from requirements, bug report summarisation, and technical documentation creation at scale.
How AI changes my QA workflow
Accelerating test case creation from requirements and specifications using LLMs — what used to take hours of manual analysis can be bootstrapped in minutes, leaving more time for the edge cases that matter.
Using AI to analyse test results and surface patterns: flaky test clusters, regression hotspots, and coverage gaps that are hard to spot manually in large suites.
Generating realistic, diverse test data at scale — user profiles, transaction sets, edge-case inputs — without maintaining brittle fixture files.
AI-assisted code review as a quality gate in CI pipelines: catching logic errors, missing assertions, and anti-patterns before they reach production.
Philosophy
"AI doesn't replace the thinking — it removes the friction so I can spend more time on what matters: strategy, edge cases, and quality that ships."