Standalone LLM-based code review tool. Paste any GitHub or GitLab URL to flag style and convention violations. Pluggable LLM backends (Claude, Gemini, Ollama, OpenAI), structured JSON output, GitHub Actions integration.

End-to-end test automation framework with self-healing locators and LLM-driven test data generation. Selenium + Cucumber base; LLM analyzes selector failures and proposes fixes; generates valid, boundary and invalid datasets per field type.

Test automation framework in Playwright using Java with BDD, parallel execution, multi-browser, and CI integration. Built as the Java counterpart to the TypeScript Playwright BDD project.

Production-grade Playwright framework with Cucumber HTML, Allure and Bootstrap reports, parallel execution, multi-browser (Chrome/Firefox/WebKit) and GitHub Actions CI.

Behave + Selenium + Requests for UI & API testing. Page Object Model, parallel execution, environment configurability, and Allure / Behave reports. Designed for clean BDD-first test authoring.

Self-healing automation platform that plugs in multiple LLM providers to recover from locator and step failures during test execution.

Containerized Selenium Grid setup for scalable, parallel cross-browser test execution. Hub + node compose for fast local and CI runs.

Node.js + Express REST API with GET / POST / PUT / DELETE endpoints. Designed as a quick bootstrap for building and testing your own RESTful service.

Shopping cart built on the MERN stack (MongoDB, Express, React, Node.js). Clone-and-run sample for learning end-to-end e-commerce flow.

Starter project for creating a custom AI agent. Useful as a teaching baseline for prompt engineering, tool use, and agent loops.