Uber builds Enhanced Agentic-RAG (EAg-RAG) to raise Genie on-call copilot answer quality to near-human precision
Genie's initial RAG implementation, when tested against 100+ golden queries on Uber's engineering security and privacy policy documents, produced responses that were incomplete, inaccurate, or failed to retrieve relevant information — falling short of the standards needed for broader Slack channel deployment. PDF loaders also failed to preserve complex table formatting, degrading downstream chunking and retrieval quality.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Enriched document ingestion
Policy documents are loaded from Google Docs via a custom Python API loader, and an LLM enriches table contents, summaries, FAQs, and keywords before chunking and indexing into a vector store.
By transitioning to an Enhanced Agentic-RAG (EAg-RAG) architecture, Uber increased the percentage of acceptable answers by a relative 27% and reduced incorrect advice by a relative 60%, enabling Genie to scale across multiple security and privacy Slack channels with a measurable reduction in on-call engineer and SME support load.
What failed first
Traditional PDF loaders (SimpleDirectoryLoader from LlamaIndex and PyPDFLoader from LangChain) stripped table formatting, disconnecting cells from their row and column context and degrading chunking and retrieval. Experiments with newer loaders including PdfPlumber, PyMuPDF, and LlamaParse could not provide a universal solution. SME-driven evaluation also took weeks per experiment, stalling iteration with only marginal accuracy gains.