RAG evaluation · Computer-vision QA · Test automation · 5+ years
I validate AI/LLM, RAG, and computer-vision systems for production reliability, and build automation frameworks (Selenium, Playwright, API) that catch regressions before release. Background spans manual, API, and automation QA across web, chatbot, and camera-based products — 1000+ cameras validated, 120+ AI benchmark cases built.
Looking for Senior QA (Manual), Automation QA, or AI/LLM Test Engineer roles. Let's connect.
- 🔭 Currently building QA-Agent — an autonomous QA copilot on the Claude Agent SDK
- 🧪 5+ years across Manual, API, Automation & AI/LLM testing
- 🏆 1000+ cameras validated · 120+ AI/RAG benchmark cases built
- 🌱 Currently deepening RAG evaluation & agentic-AI testing practices
- 📫 Reach me at krishnam.qaengineer@gmail.com
| Area | Tools |
|---|---|
| 🤖 AI / LLM & agent testing | Golden datasets · Precision/recall · Hallucination rate · Promptfoo · DeepEval |
| 📷 Computer vision | Camera / event detection · OCR test assets |
| 💬 Conversational AI | Voice-bot / chatbot validation · Intent recognition · NLP / STT |
| 🧭 Automation | Selenium WebDriver · Playwright · TestNG · Cucumber · Hybrid POM |
| 🔗 API & database | Postman · Rest Assured · SQL |
| 🔁 Process | Agile/Scrum · JIRA · CI/CD · Jenkins · Git · Docker |
QA-Agent — Autonomous QA copilot built on the Claude Agent SDK that turns specs into traceable test cases and runs real Playwright, API & database checks — not mocked.
TypeScript · Claude Agent SDK · Playwright
Promtfoo_Playwright — Evaluation pipeline for multi-agent RAG chat APIs: loads a dataset, drives the chat UI, parses streamed responses, and grades answers with semantic scoring — not just uptime checks.
JavaScript · Promptfoo · RAG eval
API-Automator-XLS — API automation framework for data-driven testing.
⭐ Show some love — if any of these QA/AI projects are useful, a star helps others find them.
📄 Resume · 🌐 Full portfolio · 📱 +91 88974 04807



