I will test your ai agent, llm app or chatbot for bugs and prompt injection
Senior SDET, Playwright and AI Chatbot Testing, 9 Years in QA
About this Gig
Shipping an AI agent, chatbot or LLM feature and not sure how it behaves when real users push it?
I'm Tajveed, a Senior SDET with 9 years in software and 6 in test automation. I spent 5 years testing AI-powered enterprise systems at Afiniti. I test AI products the same way.
What I look for:
- Broken, empty or badly formatted responses
- Context lost across turns in a conversation
- Failed or incorrect tool and function calls
- Prompt injection and jailbreak attempts on your system prompt
- Hallucination and drift away from your own knowledge base
- Edge cases: long inputs, other languages, special characters
- Timeouts, rate limits and error handling
What you get:
- A clear defect report: steps to reproduce, expected vs actual, severity, screenshots
- Playwright automation you can re-run on every release (Standard and Premium)
- CI setup so the suite runs on every push (Premium)
Stack: Playwright, Python, pytest, JavaScript, TypeScript, Node.js, Postman.
Send your app link or repo before ordering and I'll tell you what's worth testing and which package fits.
Testing application:
Web application
Development technology:
JavaScript
•
Node.js
•
Python
•
React
•
TypeScript
Device:
PC
•
Mac
•
Linux
•
iPhone
•
Android mobile phone
My Portfolio
FAQ
Can you test an AI agent that calls tools or APIs?
Yes. I test tool and function calls, API handoffs, retries and timeouts, and report exactly which call failed and with what input. Works with OpenAI, Gemini, Claude and custom endpoints.
What do you need from me to get started?
A link to your app, chatbot or staging environment, plus test access if it needs login. If you have a system prompt, knowledge base or API docs, share those too. Not required, but it makes the testing sharper.
Do I get to keep the automated tests?
Yes. Standard and Premium include the full Playwright test code, with a README so your team can run it locally or in CI. Basic is a manual test pass and comes with the defect report only.

