Traditionally, course materials are presented in a linear manner, usually through lectures conducted across the duration of a semester. This method of delivery can hinder students' understanding of advanced topics as it struggles to facilitate connections across different lectures. To address these issues, course-KG provides an automated system which processes lecture transcripts and organizes key information into a knowledge graph (KG). This medium provides students a contextually rich representation of course materials by capturing the semantic relationships between concepts. This is accomplished by leveraging state of the art advancements in Natural Language Processing (NLP) and Large Language Models (LLMs). Through the application of various pre/post processing tasks including text splitting, summarization, and multi-prompting, course-KG’s modular framework is able to produce reasonable knowledge graphs representative of course content. The platform also provides an intuitive interface where users can interact with generated KGs and make any necessary modifications. Despite the promising results however, issues such as hallucinations, bipartite subgraphs, and extended task completion time, compromise the quality of KGs produced. Future enhancements such as overlapped text splitting, LLMs with larger context windows, and course-specific prompting, can address these weaknesses and enhance this system’s KG generation capabilities.