Using Large Language Models to Provide Commercial Insights to a Leading Pharmaceutical Company

  • Published: March 28, 2024



Learn how our client benefitted from a custom-built large language model solution that was able to extract insights from extensive unstructured data through an intuitive user interface.

Unstructured Data – The Bane of Commercial & Medical Teams

The client’s commercial and medical teams needed access to fast and actionable insights to function effectively. This proved challenging when the backend data was unstructured and did not have an easy-to-use data extraction interface.


Reveal’s Approach

Reveal engaged in in-depth discussions with the commercial and medical teams to understand their requirements. Multiple large language models were applied to enable semantic search, topic modeling, sentiment analysis, key phrase extraction, and text summarization of various data sources. A platform was engineered on AWS using OpenSearch with an intuitive user interface and easy to understand dashboards. The UI was custom-built to address the specific workflows of the commercial and medical teams.


Business Impact

The platform was well-received by its intended users. They were greatly pleased by the –

  1. Simplified access to unstructured data
  2. Responsiveness of the search functionality
  3. Easy-to-use UI