Grep or Embeddings? Agentic Search Over Company Documents — George He, LlamaIndex

Grep works great on a codebase. Try it on millions of PDFs, slides and scanned schematics. George He, head of engineering at LlamaIndex, explains when agents need pre-indexed search and when plain file traversal is enough. Claude Code skipped embeddings because code is small, text-based, local and full of breadcrumbs. Company data is huge, messy, multimodal and permissioned. He shows how to give an agent a harness of tools so it can choose: hybrid retrieval as a compass, then listing, metadata filtering, grep and reading to dig in, plus parsed text and page screenshots for complex documents. He covers parsing with LlamaParse and the open-source LiteParse, permissions and freshness in production, and the ingest-parse-index pipeline. Then he demos an agent building a cash-flow table from 135 Alphabet financial filings, grounded in the source files. In this talk: • Why Claude Code's grep approach works for code but breaks on company documents • Hybrid search as a compass, then grep and read to ground the answer • Parsing complex PDFs, tables and schematics, with page screenshots • Production concerns: multi-tenancy, permissions and data freshness SPEAKER George He, Head of Engineering, LlamaIndex LinkedIn: https://www.linkedin.com/in/georgehe4/ GitHub: https://github.com/georgehe4 LINKS LiteParse (GitHub): https://github.com/run-llama/liteparse LlamaParse docs: https://developers.llamaindex.ai/llamaparse/ LlamaIndex: https://www.llamaindex.ai LlamaIndex on X: https://x.com/llama_index CHAPTERS 0:00 Intro 1:12 When do agents need vector search? 2:17 Why Claude Code skipped embeddings 3:27 What makes code easy 3:57 Company documents are different 5:17 It depends on scale 6:32 Bigger context isn't free 7:12 Sub-agents get expensive 8:22 Let the agent choose 8:57 Five tools for document search 10:52 Tool 1: hybrid search 11:57 Tools 2–4: list, filter, grep 12:42 Tool 5: reading complex documents 13:27 LlamaParse and LiteParse 14:22 Measuring parse quality 15:17 Permissions and freshness 15:57 Ingest, parse, index 16:42 Storage and metadata filtering 17:47 Demo: Alphabet financial filings 20:07 Demo: grep and read remote files 20:47 Demo: building a cash-flow table 22:52 Takeaway Recorded at the AI Engineer World's Fair 2026 in San Francisco. Subscribe for more talks from the engineers building with AI. AI Engineer: https://ai.engineer YouTube: https://www.youtube.com/@aiDotEngineer X: https://x.com/aiDotEngineer LinkedIn: https://www.linkedin.com/company/aidotengineer/ #RAG #DocumentAI #AIEngineer