Building testing architectures that think — bridging the gap between Quality Engineering and Generative AI.
- Independent R&D (Present): Architecting GenAI-integrated QA tools—exploring self-healing locators, LLM evaluation, and prompt-based test generation. Available for Remote · Hybrid · Relocation roles.
- Previous Impact (CloudKaptan): Engineered a 300+ test case WebdriverIO suite, reducing regression time by 40% and pipeline execution from 8 hours to 1.5 hours. Served as technical SPOC for global clients and conducted 50+ technical interviews for QA/QE hiring.
| Project | Architectural Overview | Tech Stack |
|---|---|---|
| Self-Healing Playwright Framework | Integrates Gemini 2.5 Flash to automatically diagnose broken locators on test failure and generate reviewable fixes. | TypeScript, Playwright, Qase, GitHub Actions |
| LLM-as-a-Judge | Custom evaluator scoring GPT-OSS responses across 5 quality dimensions (accuracy, relevance, completeness, tone, hallucination). | TypeScript, Vitest, Gemini API |
| Unified API Testing Client | Multi-protocol testing architecture unifying REST, GraphQL, and WebSocket under a single strongly-typed framework. | TypeScript, Vitest, Faker.js, GitHub Actions |
| TestGenie | AI-powered test case generator converting feature descriptions into Functional, Security, Boundary, and UX tests with direct export. | Gemini API, TypeScript, Vercel |
- Automation & Frameworks: Playwright, WebdriverIO, Selenium, Vitest, Mocha
- Languages & DevOps: TypeScript, JavaScript, Java, SQL, GitHub Actions, Azure DevOps, Locust
- API Testing: Postman, Axios, GraphQL, WebSocket
- GenAI / AI Testing: LLM Evaluation, Hallucination Detection, Prompt-based Test Generation, Self-Healing Locators
Driving Quality Engineering Excellence — Mentorship, Pipeline Optimization, and Advanced AI Integrations.