How to Build AI Agents by Augmenting LLMs with Codified Human Expert Domain Knowledge? A Software Engineering Framework

TL;DR

This paper presents a framework for embedding human expert knowledge into AI agents to improve visualization tasks, enabling non-experts to achieve expert-level results across engineering domains.

Contribution

It introduces a novel software engineering framework that captures and codifies human domain knowledge into AI agents, enhancing visualization quality and accessibility.

Findings

206% improvement in output quality

Agent achieved expert-level ratings in all scenarios

Lower variance and higher code quality than baseline

Abstract

Critical domain knowledge typically resides with few experts, creating organizational bottlenecks in scalability and decision-making. Non-experts struggle to create effective visualizations, leading to suboptimal insights and diverting expert time. This paper investigates how to capture and embed human domain knowledge into AI agent systems through an industrial case study. We propose a software engineering framework to capture human domain knowledge for engineering AI agents in simulation data visualization by augmenting a Large Language Model (LLM) with a request classifier, Retrieval-Augmented Generation (RAG) system for code generation, codified expert rules, and visualization design principles unified in an agent demonstrating autonomous, reactive, proactive, and social behavior. Evaluation across five scenarios spanning multiple engineering domains with 12 evaluators demonstrates 206% improvement in output quality, with our agent achieving expert-level ratings in all cases versus baseline's poor performance, while maintaining superior code quality with lower variance. Our contributions are: an automated agent-based system for visualization generation and a validated framework for systematically capturing human domain knowledge and codifying tacit expert knowledge into AI agents, demonstrating that non-experts can achieve expert-level outcomes in specialized domains.