Digital Adoption Platforms (DAPs) have become essential tools for helping employees navigate complex enterprise software such as CRM, ERP, or HRMS systems. Companies like LemonLearning have shown how digital guidance can reduce training costs and accelerate onboarding. However, building and maintaining these interactive guides still requires extensive manual effort. Leveraging Large Language Models as virtual assistants is an appealing alternative, yet without a structured understanding of the target software, LLMs often hallucinate and produce unreliable answers. Moreover, most production-grade LLMs are black-box APIs, making fine-tuning impractical due to the lack of access to model weights. In this work, we introduce a Graph-based Retrieval-Augmented Generation framework that automatically converts enterprise web applications into state-action knowledge graphs, enabling LLMs to generate grounded and context-aware assistance. The framework was co-developed with the AI enterprise RAKAM, in collaboration with Lemon Learning. We detail the engineering pipeline that extracts and structures software interfaces, the design of the graph-based retrieval process, and the integration of our approach into production DAP workflows. Finally, we discuss scalability, robustness, and deployment lessons learned from industrial use cases.

翻译：数字采用平台已成为帮助员工驾驭复杂企业软件（如CRM、ERP或HRMS系统）的重要工具。LemonLearning等公司已展示数字指导如何降低培训成本并加速员工入职。然而，构建和维护这些交互式指南仍需大量人工投入。利用大型语言模型作为虚拟助手是一种有吸引力的替代方案，但若缺乏对目标软件的结构化理解，LLM常产生幻觉并给出不可靠的答案。此外，大多数生产级LLM为黑盒API，由于无法访问模型权重，微调实际不可行。本研究提出一种基于图的检索增强生成框架，可自动将企业Web应用程序转换为状态-动作知识图谱，使LLM能够生成基于事实且上下文感知的辅助内容。该框架是与AI企业RAKAM合作，协同Lemon Learning共同开发的。我们详述了提取和结构化软件界面的工程流程、基于图的检索过程设计，以及将该方法集成至生产级DAP工作流的方案。最后，我们基于工业用例讨论了可扩展性、鲁棒性及部署经验教训。