You get a report, not a shrug.
- What was tested. Every screen and flow that was covered.
- How it was tested. The testing types used and the steps followed.
- What broke. Each bug written so an owner can follow it, not only a developer.
- What to change. A recommended fix for each one, in order of priority.
Checkout: "Pay now" stays disabled after a saved card is edited
- Steps
- Add an item, go to checkout, edit the saved card, return to the payment step.
- Expected
- The button becomes active once the card details are valid.
- Actual
- The button stays greyed out until the page is refreshed.
- Fix first
- Re-run form validation when the card is updated.
Find the problems while they are still cheap to fix.
Careful manual testing through the eyes of a user, backed by automated tests that keep checking after every release.
- Manual testing
Working through your web app screen by screen to find bugs, broken flows and confusing steps.
- Test plans and cases
Written scenarios and step by step cases that cover every module, so nothing is tested by guesswork.
- Test automation
End to end tests in Playwright and Cypress that run again whenever the product changes.
- API testing
Checking that the systems behind the app return the right data and handle errors properly.
- Accessibility testing
Reviewing and fixing products so that people with disabilities can use them.
- Usability checks
I trained as a product designer, so the report covers what feels wrong as well as what is broken.
The loop.
Every engagement runs the same four steps, round and round, until the product is ready.
Test
Your app is tested by hand and with code, against an agreed list of what matters.
Report
You receive a structured report: testing types used, steps followed, bugs found.
Fix
Each bug comes with a recommended change, so your developers know where to start.
Retest
The fixes are checked again, and the loop closes when the checks pass.
Recent work.
Designed 56 test scenarios across 11 application modules, with automated API tests and shared test coverage across the product.
Testing the app directly for its owner through the user interface. Reports state the testing types used, the step by step tests, the bugs found and the recommended adjustments.
Building the authentication, income and dashboard screens in TypeScript, and testing them with sample data while the backend is in progress.
One public project every week that solves a real problem. The rotation covers Playwright testing, Cypress testing, API testing and accessibility fixes, with beginner friendly issues so other developers can contribute.
Tools: Playwright, Cypress, API testing, accessibility testing, GitHub, Figma, Claude Code.
I also build the boring bits so you don't have to.
AI workflows built in n8n, and tested with the same care as any product: every workflow is checked to confirm it does what it promises.
What I build
- Workflow automation. Connecting your apps so routine tasks run on their own.
- AI agents. Single and multi agent systems, including agents that use MCP to reach other tools.
- Voice and assistant tools. AI assistants people can talk to.
- Prompt design. Prompts written and chained for reliable output with fewer rounds of fixes.
Projects I have built
- Two agent sales system. Two AI agents working together through MCP.
- Voice assistant. A working assistant you can speak to.
- ETF portfolio rebalancer. An automated rebalancing workflow.
- AAC Quick Phrase Assistant. Helps children with speech, vision or hearing challenges communicate using ready made phrases.
- Claude Code skill pack. Twelve reusable AI skills for design, building, testing and automation tasks.
These are my own working projects, not client engagements.
Tell me what you are building.
Share a few details and I will reply by email. Want a rough idea of cost first? See packages and prices. Not sure you are ready? Take the 2-minute launch check.
Email[email protected]
LinkedInlinkedin.com/in/oluwatobiloba-bilewu
GitHubgithub.com/Bilex95
AvailabilityRemote, worldwide