RAG in the NASA Domain - Technical & Lessons Learned Datasets

The datasets are part of an Retrieval-Augmented Generation (RAG) system specialized in aerospace knowledge retrieval and generation. The Technical Documents Dataset contains over 8,000 NASA technical reports and publications extracted from the NASA Technical Reports Server (NTRS), including conference papers, contractor reports, presentations, and other technical documentation spanning decades of aerospace research and development.

The Lessons Learned Dataset comprises more than 1,800 structured lessons-learned documents scraped from the NASA Lessons Learned Information System (LLIS), providing critical insights from past missions, failures, and successes.