Method
Agentic RAG
RAG where an agent decides when and how to retrieve, iteratively.
9 cases from 9 companies apply Agentic RAG.
Typical tasks: Build an assistant over internal docs, Answer questions over data, Reduce ML cost, Rank search results and 3 more. Companies: Databricks, Sber, VTB, LinkedIn, Atlassian, X5 Tech and 3 more. Often combined with: BM25, LLM-as-a-judge, RAG, Hybrid search, Synthetic data generation (LLM) and 34 more.
| Behind the curtain: Key learnings from building Genie | general | 2026-06 | |
| How a 4B Model Beat GPT at Agentic Retrieval (And How to Train One on a Neon Postgres Corpus)CA | CA | general | 2026-06 |
| The Search of the Future | general | 2026-05 | |
| Building a Battle-Tested Open-Source Agentic RAG | general | 2026-05 | |
| RAG for Complex Situational Questions | banking | 2026-05 | |
| LinkedIn Cognitive Memory Agent: long-term memory for personalizing LLM agents | saas-b2b | 2026-03 | |
| How Rovo Deep Research works | saas-b2b | 2025-08 | |
RAG Tricks: Practical Steps for Corporate Document RAG![]() | ![]() | general | 2024-05 |
| Building DoorDash's product knowledge graph with large language models | food-delivery | 2024-04 |
