For the enterprises of today, there is an increased focus on continuous learning and skill development of the workforce to drive sustainable growth. To achieve this, enterprises need to identify key areas of focus and expertise for their businesses, keep an eye on the latest developments in those fields, and ensure that they have the right learning material available for the workforce to help them upskill. This necessitates rapid production and dissemination of learning material through the Learning Experience Platforms (LXPs) of the enterprises across multiple topics, and this is where the recent developments in the fields of Generative AI could come in handy. This work focuses on the implementation of a generative AI-augmented authoring tool in the LXP for a large enterprise, where metadata for the learning materials are generated with the help of an open source LLM that is fine-tuned with organizational knowledge. The LXP then takes the human-in-the-middle approach to get the metadata reviewed by experts before publishing the learning material. This work discusses the challenges of using a pre-trained LLM for this activity, and how fine-tuning LLMs specific to the task could help improve the quality and performance. It also validates the generated metadata using well-known evaluation metrics to highlight the usefulness of the approach to establish how the fine-tuned model performs better than larger base models.