AI-DRIVEN AGENTIC FRAMEWORK FOR ENTERPRISE SYSTEM TROUBLESHOOTING WITH ENHANCED RAG MODELS

Abstract

Technical troubleshooting in enterprise settings frequently requires navigating various, heterogeneous data 
sources to effectively resolve intricate issues. This paper introduces an innovative agentic AI solution based on a 
Weighted Retrieval-Augmented Generation (RAG) Framework designed for enterprise technical troubleshooting. 
By dynamically weighting retrieval sources, including product manuals, internal knowledge bases, FAQs, and 
troubleshooting guides according to query context, the framework prioritizes the most pertinent information. It 
prioritizes product manuals for SKU-specific inquiries while integrating general FAQs for more extensive 
concerns. The system utilizes FAISS for efficient dense vector search, along with a dynamic aggregation 
mechanism to integrate results from various sources seamlessly. A LLaMA-based self-evaluator guarantees the 
contextual precision and assurance of the produced responses prior to their dissemination. This iterative process 
of retrieval and validation improves precision, diversity, and reliability in response generation.Initial assessments 
of extensive enterprise datasets indicate the framework's effectiveness in enhancing troubleshooting precision, 
decreasing resolution durations, and accommodating diverse technical challenges. Future research intends to 
improve the framework by incorporating sophisticated conversational AI functionalities, facilitating more 
interactive and intuitive troubleshooting experiences. Efforts will concentrate on enhancing the dynamic 
weighting mechanism via reinforcement learning to optimize the relevance and accuracy of retrieved information. 
By integrating these advancements, the proposed framework is set to transform into a holistic, autonomous AI 
solution, revolutionizing technical service workflows within enterprise environments