Method

Few-shot prompting

Giving examples in the prompt, incl. dynamic few-shot retrieval.

36 cases from 24 companies apply Few-shot prompting. 4 were checked in an online A/B test; 13 report results in production.

Typical tasks: Evaluate models, Rank search results, Label data with LLMs, Build an assistant over internal docs and 10 more. Companies: Yandex, Instacart, DoorDash, Airbnb, Yelp, Etsy and 18 more. Often combined with: LLM-as-a-judge, RAG, Prompt engineering, Supervised fine-tuning, Guardrails and 87 more.

CASTLE: training natural language search before the first real querytravel2026-10
Building Gavel: AI evaluation infrastructure across Eternalfood-delivery2026-10
Eval-driven development: Lessons from evaluating GenAI at scaletravel2026-08
Automating Search Relevance Assessment at Scale with LLM-as-a-Judgee-commerce2026-08
Encoding Your Domain Expert: The Context Layer Behind Spotify's Data Assistantmedia-streaming2026-06
How LLMs Take Clustering to a New Levelretail2026-05
An Automated Request Handling Service for SMBsbanking2026-05
Developing an Aspect-Based Sentiment Analysis System for Online Course Reviews Using LLMsedtech2026-05
Building Biz Ask Anything: From Prototype to Productmarketplace2026-03
Beyond the dashboard: how BlaBlaCar PMs use AI to self-serve datamobility2026-03
How Etsy Uses LLMs to Improve Search Relevancemarketplace2026-01
Creating a domain-specific NL-to-SQL MCP serversocial2025-11
Building The Intent Engine: How Instacart is Revamping Query Understanding with LLMse-commerce2025-11
Automating the Labeling of News Queries with RAGsearch-engine2025-11
Understanding Etsy's Vast Inventory with LLMsmarketplace2025-10
Using LLMs to infer grocery preferences from DoorDash restaurant ordersfood-delivery2025-09
Scaling Catalog Attribute Extraction with Multi-modal LLMse-commerce2025-08
Advancing Menu Content with AI: How DoorDash uses AI to generate menu descriptionsfood-delivery2025-06
Introducing a RAG-based bot to streamline CSsocial2025-06
Estimating Generative Model Confidence in Data Labelinggeneral2025-05
Catch Me If You Can: LLM Quality Control for Customer Servicegeneral2025-05
Building RAG-Like Systems for Few-Shot Intent Classification in Narrow Domainsbanking2025-05
Product Applications of LLMs: Building an AI Copilot for Editors and Speeding Up Time to Marketbanking2025-05
Cycle Paths, Shashlik and Sakura: How Yandex Maps Built Seasonal Leisure Recommendationsgeneral2025-05
5 Stops on the Way to LLM in Your Productiongeneral2025-05
Search Query Understanding with LLMs: From Ideation to Productionmarketplace2025-02
Automated GenAI-driven search quality evaluationsocial2024-12
LLMs vs. Annotators: How to Get High-Quality Text Labelingadtech2024-11
Supercharging Discovery in Search with LLMse-commerce2024-09
Using Predictive and Gen AI to Improve Product Categorization at Walmarte-commerce2024-07
A Personal Mentor Powered by GPTedtech2024-05
LLMs, NER and Regular Expressions for Resume Parsinggeneral2024-05
RAG Tricks: Practical Steps for Corporate Document RAGgeneral2024-05
Building DoorDash's product knowledge graph with large language modelsfood-delivery2024-04
Navigating Fair Housing Guardrails in LLMsmarketplace2024-01
Text interfaces for data—general2023-05