Proceedings of the International scientific and practical conference ―New York Global Science Conference 2026‖ (May 18-20, 2026) / Publisher website: www.naukainfo.com. – New York, USA, 2026. - 322 p.

127 COMPUTER AND SOFTWARE ENGINEERING UDC 004.415.53:004.89 Kutaiev Serhii Valentynovych senior lecturer Lviv State University of Internal Affairs Lviv, Ukraine INTELLIGENT DEBUGGING AND AUTOMATED ROOT CAUSE ANALYSIS THROUGH LLMS INTEGRATION IN COMPLEX E2E SCENARIOS Abstract. Modern E2E (End-to-End) automation frameworks operating across heterogeneous environment layers generate vast amounts of disparate diagnostic data upon test failure. Isolating the actual root cause of a defect across desktop interfaces, network logs, and backend database states remains a highly time-consuming task for QA engineers. This paper introduces a methodology for intelligent debugging and automated Root Cause Analysis (RCA) leveraging Large Language Models (LLMs). The proposed architectural pattern involves capturing contextual execution data - such as Appium stack traces, REST API request-response payloads, and OS-level exceptions - and passing it to an orchestrated LLM agent. The study demonstrates how generative AI can semantically correlate these artifacts to instantly classify the failure (e.g., application bug, environment instability, or test script flakiness), thereby reducing debugging latency and accelerating the feedback loop in CI/CD pipelines. Keywords: intelligent debugging, Root Cause Analysis, LLMs, E2E testing, automated diagnostics, Appium, REST API.