# Yelp Published articles for Yelp. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Yelp Combines Native and Server-Driven UI for Its AI Assistant DevFeed: [Yelp Combines Native and Server-Driven UI for Its AI Assistant]() Original publisher: [Read original article]() Author: Leela Kumili Published: 2026-11-06T14:25:00Z Content type: article Language: en Sources: [InfoQ]() Topics: [server-driven-ui](), [AI Chat]() Tags: [ai](), [ai-architecture](), [ai-assistant](), [architecture](), [architecture-design](), [backend](), [challenges](), [chaos](), [chat](), [chatbots](), [configuration](), [cross-platform](), [data](), [development](), [event-driven-architecture](), [faster](), [framework](), [large-language-models](), [mobile](), [mobile-development](), [native](), [news](), [pilot](), [platform](), [rendering](), [server](), [server-driven-ui](), [server-side-rendering](), [ui](), [yelp](), [yelp-assistant-hybrid-ui]() ### AI overview Yelp's AI Assistant uses CHAOS for server-driven conversation rendering while keeping navigation and message input native. The backend sends a view configuration when a chat starts, then sends conversation events that update the existing view without refetching that configuration. Events can carry persistent messages, temporary typing indicators and quick replies, links, or actions such as disabling input when a project request is complete. This separation lets supported presentation changes reach users without a new app release and helped Yelp reuse its Android Assistant for the Biz Ask Anything pilot. The client still handles message state and interactions; recoverable errors trigger retries, while non-recoverable failures disable input and ask users to try again later. Yelp reports coordination and learning costs, and shared implementations may still behave differently across platforms. ### Source excerpt Yelp combines native mobile controls with server-driven conversation rendering in its AI Assistant using the CHAOS framework. The architecture separates view configuration from conversation data, supports backend-driven chat interactions, and enables faster UI updates without new app releases. Yelp also discusses cross-platform challenges and reuse for its Biz Ask Anything pilot. By Leela Kumili ## Should your site let AI agents in? Yelp charges, JPMorgan says no. DevFeed: [Should your site let AI agents in? Yelp charges, JPMorgan says no.]() Original publisher: [Read original article]() Author: WorkOS Published: 2026-10-22T00:00:00Z Content type: opinion Language: en Sources: [WorkOS]() Topics: [AI Bots](), [API Monetization](), [API Business Models](), [Dynamic authorization rules](), [Shopify]() Tags: [actions](), [admission](), [agent](), [agentic](), [agents](), [ai](), [ai-agents](), [api](), [bot](), [bot-detection](), [charges](), [contracts](), [credentials](), [customer](), [data](), [shopify](), [yelp]() ### AI overview The article argues that sites should manage AI agent access as a commercial product decision, using operator-specific credentials with defined scopes, rate limits, agreements and revocation. Bot detection cannot distinguish a licensed agent from a scraper, so it can block approved traffic or admit unwanted traffic. Examples show different approaches: Yelp licenses access; JPMorgan rejects agents using customers' banking credentials while supporting secure, authorized sharing; Shopify lets merchants control checkout and catalog access by AI channel. Agent identity protocols can provide a stable identifier, but do not establish who operates an agent or what it may do. The article recommends deciding which actions to offer, at what limits and under what liability terms, and setting a process for resolving partner blocks. It notes that paid admission risks excluding independent developers, making a usable free tier important. ### Source excerpt Yelp sells agent access, JPMorgan refuses it, Shopify ships it as a toggle. Admission is a commercial call, and bot rules are the wrong place to enforce it. ## Building React Apps with GraphQL, Graphcool & Apollo DevFeed: [Building React Apps with GraphQL, Graphcool & Apollo]() Original publisher: [Read original article]() Author: Graphcool Published: 2017-06-07T00:01:38Z Content type: tutorial Language: en Sources: [Apollo Blog]() Topics: [GraphQL](), [React](), [apollo-client](), [Instagram]() Tags: [api](), [apis](), [apollo](), [apollo-client](), [backend](), [cli](), [community](), [facebook](), [github](), [graphql](), [instagram](), [react](), [twitter](), [yelp]() ### AI overview The tutorial walks through setting up a React Instagram app with Apollo Client and a Graphcool GraphQL backend. It defines the app's data model in GraphQL SDL, uses the Graphcool CLI to create a backend from a remote schema, and points to a GraphQL Playground for exploring queries and mutations. The starter app is already implemented; setup consists of configuring its client with the endpoint Graphcool provides. The excerpt says Graphcool is intended for a production-ready backend, while Apollo Launchpad and graphql-up are presented as easy ways to spin up APIs but not for production use. It begins to explore the app components, but the supplied text ends mid-sentence. ### Source excerpt It's been less than two years since Facebook first open-sourced GraphQL as a new way of exposing data from an API. In that short period, it's already revolutionized the API space and many major companies, like GitHub, Twitter and Yelp, have started using it in production. In this article, you'll learn how to quickly get started with React & GraphQL by using Apollo Client and Graphcool to building a simple Instagram app. You can find the final version of the code on GitHub.