Method

Dense retrieval

Retrieving with bi-encoder text embeddings instead of keywords.

28 cases from 21 companies apply Dense retrieval. 6 were checked in an online A/B test; 6 report results in production.

Typical tasks: Rank search results, Build an assistant over internal docs, Retrieve and rank from millions of items, Evaluate models and 6 more. Companies: DoorDash, LinkedIn, Atlassian, Yandex, Databricks, Meta and 15 more. Often combined with: Hybrid search, LLM-as-a-judge, Cross-encoder reranking, BM25, RAG and 68 more.

Inside Vera, DoorDash’s Data Agentfood-delivery2026-09
Rebuilding LinkedIn's Follows Recommendations with LLM-Based Semantic Retrieval and Rankingsocial2026-08
Effective document management and retrieval for generative AIbanking2026-06
How Mastercard Turns Transaction Data into Trusted Merchant Insights with AIfintech2026-06
Building a Battle-Tested Open-Source Agentic RAGgeneral2026-05
Modernizing the Facebook Groups Search to Unlock the Power of Community Knowledgesocial2026-04
Using LLMs to build content embeddings for search and recommendationsfood-delivery2026-04
Advancing semantic search for millions of Rovo userssaas-b2b2026-03
How Rovo solves search challenges through entity linkingsaas-b2b2026-01
When GenAI Meets Personalization: Powering DoorDash's next-generation homepage experiencefood-delivery2025-12
Greeting + Does RAG Even Work?fintech2025-11
Building the next generation of job search at LinkedInmarketplace2025-05
How LLM Automates Work at Avitoclassifieds2025-05
YandexGPT as a RAG assistant for tech support: solving problems fastergeneral2025-05
Make Retrieval Smarter: How We Built Smart Search over a Document Basegeneral2025-05
LLM-Based Personalization of Alice's Conversations with Usersgeneral2025-05
Practical Text-to-SQL for Data Analyticssocial2024-12
Optimizing search relevance at Instacart using hybrid retrievalmarketplace2024-09
Deploying RAGs in production — Part 1retail2024-07
GigaSearch, or a Search System on GigaChatsearch-engine2024-05
Marketplace Product Matching in Detail: Models, Approaches and Tips for Building an Effective Matcher for Tens of Millions of Productsmarketplace2024-05
RAG as a Flexible Method for Creating Language Assistantsgeneral2024-05
How We Got RAG Running in Productiongeneral2024-05
How to Build Multilingual Product Matching Without Dataretail2024-05
RAG Tricks: Practical Steps for Corporate Document RAGgeneral2024-05
Semantic Search at Zendesksaas-b2b2024-03
Introducing Natural Language Search for Podcast Episodesmedia-streaming2022-03
Kroomsa: A search engine for the curiousgeneral2022-02