NASA People Graph: AI-powered Workforce Analytics

NASA Uses Graph Databases and LLMs to Enhance People Analytics

NASA employs graph databases and LLMs to analyze employee data, identify experts, and build teams. This system uses Memgraph, Ollama, and various data sources to extract skills and detect project overlaps, allowing for a RAG-based chatbot and plans to scale the graph.

1. NASA's People Graph is a Cutting-Edge Application of Graph Databases: It moves beyond traditional relational databases to capture complex relationships between employees, skills, and projects, showing the potential for advanced people analytics.
2. GraphRAG Enables Natural Language Queries for Expertise Discovery: The use of a RAG-based chatbot allows users to query the knowledge graph in natural language, making it easier to find subject matter experts and organizational insights.
3. NASA's Infrastructure Prioritizes Security with On-Prem LLMs: Running the LLM server on NASA's secure internal AWS cloud (Olama on EC2) demonstrates a commitment to data security and control, a key concern for sensitive information.
4. Skill Extraction is Streamlined with LLMs: NASA leverages LLMs (like those accessible via Ollama) to automate skill extraction from resumes and other data sources, reducing manual effort and improving accuracy.
5. Flexibility and Iteration are Key to Graph Development: NASA's approach emphasizes starting with known relationships and dynamically adding new ones, like project similarity, showcasing the adaptability of knowledge graphs.