Engineering Copilot

BMW and Sopra Steria built an engineering copilot with Neo4j and RAG to support functional safety, requirements management, and system modeling for complex E/E and software-development data.

Data requirements

• E/E data (tabular) — The copilot works with complex E/E engineering data.
• software-development data (text) — The copilot works with software-development artifacts and related information.

AI methods and techniques

• Retrieval-Augmented Generation (RAG) — Used to retrieve relevant context from connected engineering data before generating outputs.
• knowledge graph — Used as the structured layer for connected E/E and software-development data.

Industries

Real-world evidence

14 documented case studies on record.

Companies using this: adidas, Amazon, Boeing, Deloitte, Feedzai, Honeywell, Lockheed Martin, Mitsubishi Electric, QuEra, Rolls-Royce, Shopify, Snap, Volvo, BMW.