Marius Argatu

• ▹ Own the quality strategy within my feature team: what to test, at what level, and why. Developers write their own unit, integration, and system tests while I advise on coverage gaps and where tests belong.
• ▹ Maintain the system test framework and review developer-written tests, giving feedback on coverage and test design.
• ▹ Shift-left: authored test scenarios during the LLD design phase as the spec for developer-written tests, embedding quality gates early in the lifecycle.
• ▹ Help test new features to shorten the feedback cycle before handoff to the broader QA team.
• ▹ Use agentic workflows day to day: MCP as a knowledge base, slash commands for bug triage, agents for test maintenance.
• ▹ Built an internal LLM-powered debugging agent in Python that correlates failing test output with Datadog traces, the pytest stacktrace, and recent Jira changes, then loops until the test passes or a real bug surfaces. Debug time from hours to minutes.
• ▹ Owned regression planning and release-gate decisions for the payments domain.

RAG testing, framework ownership, and platform-wide test architecture.

• ▹ Built LLM evaluation for a RAG assistant answering researcher questions against FDA/EMA guidance and tenant submissions. Separated retrieval (hit rate, precision@k) from generation (faithfulness, answer relevancy) via RAGAS; golden datasets in Langfuse, eval runs in CI.
• ▹ Model-based testing on the permissions matrix using xState: ~1,000 scenarios generated from a single state-machine definition, tracking the product as permissions evolve with no manual rewrite.
• ▹ Built a Playwright + TypeScript test platform with POM and fixture-based tenant/role isolation, adopted across multiple functional teams.
• ▹ Built an AST-based migration tool that translated a legacy KarateDSL suite into equivalent TypeScript + Playwright code, removing a manual rewrite of thousands of lines.
• ▹ Framework SME: onboarding, test-pattern reviews, and coordination on cross-cutting changes including an auth-system rewrite.

Python + Pytest API test framework with dynamic data generation.

• ▹ Built a custom Pytest plugin that runs dependent API tests concurrently via DAG-resolved ordering, cutting runtime while preserving test correctness.
• ▹ Patched and rebuilt QT WebEngine's Chromium fork to expose the remote debugging port, replacing Appium with Selenium via CDP. Per-test effort dropped from 4 days to 1.

• ▹ Designed a cross-platform automation framework covering an SPA and a Windows desktop client using the Screenplay pattern with Playwright and WinAppDriver: one mental model, two surface areas.
• ▹ Delivered an E2E suite that meaningfully reduced manual regression on the historically under-automated desktop side.
• ▹ Maintained and optimized Jenkins pipelines for test execution and reporting across environments.
• ▹ Coordinated day-to-day activities for a small QA team.

• ▹ Led design of a Pytest-based automation framework for complex e-commerce and payment workflows; defined standards still in use.
• ▹ Cut test execution time substantially via parallel execution and script optimization.
• ▹ Built a database-diff tool that captured state changes triggered by UI actions, letting QA snapshot and restore environments locally. Removed hours of manual SQL from the team's weekly workload.

• ▹ Shipped RESTful APIs in Python + Flask; validated performance under load with K6; maintained CI/CD on Docker + Kubernetes.

• ▹ Owned integration testing of REST services end to end; organized daily QA across web, desktop, and mobile.

• ▹ Authored and evolved test cases against changing requirements; ran regression cycles; partnered with devs, producers, and designers on defect resolution.

• ▹ End-to-end feature testing for AAA game titles across platforms.