--- title: Inside-Out AI: Rebuilding Airbnb Behind the Scenes and... description: Airbnb CTO Ahmad Al-Dahle, formerly head of generative AI at Meta, describes an 'inside-out AI' strategy: using AI internally to speed product development,... canonical: https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51 copyright: © 2026 Daily Dev Ltd. terms: https://daily.dev/tos/ twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience | daily.dev og:description: Airbnb CTO Ahmad Al-Dahle, formerly head of generative AI at Meta, describes an 'inside-out AI' strategy: using AI internally to speed product development,... og:url: https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51 og:image: https://api.daily.dev/og/posts/tnLC2rs51.png og:image:alt: Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience **[Latent Space](https://daily.dev/sources/latentspace)** · 8 min read · 1 upvotes · 0 comments ## Summary Airbnb CTO Ahmad Al-Dahle, formerly head of generative AI at Meta, describes an 'inside-out AI' strategy: using AI internally to speed product development, then deploying the same capabilities to transform guest-facing experiences. He cites metrics including 60% AI-authored code, nearly 80% more shipped features year over year, and 1.6x higher pull-request throughput per engineer. Customer support AI now resolves roughly half of tickets, built via synthetic data testing before production rollout. An internal context graph called Everest helped Airbnb launch grocery delivery and airport pickup services quickly, with airport pickups taking about six weeks versus eight to nine months for groceries. Airbnb runs a multi-model strategy, post-training open models for specific use cases while using frontier models for coding and smaller specialized models for latency-sensitive search. The company is moving toward asynchronous agents that triage on-call incidents automatically, and Al-Dahle stresses organizing teams around outcomes rather than features while ensuring junior engineers still develop engineering judgment despite AI assistance. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What percentage of Airbnb's code is now written by AI? About 60% of Airbnb's code is AI-authored. This shift accompanied roughly 80% more features and improvements shipped year over year, and a 1.6x increase in pull-request throughput per average engineer, driven by moving product, design, and engineering teams directly to working with prototypes instead of handoff-heavy processes. _Engineering leaders tracking AI's real impact on software delivery follow case studies like this on daily.dev._ ### How much of Airbnb's customer support is handled by AI agents? Roughly half of Airbnb's support tickets are resolved purely by AI, a figure close to the nearly 45% cited in the company's Q2 2026 financial results. The system is built by generating a large battery of synthetic data to test the agent thoroughly before production, while certain sensitive cases like safety issues are deliberately routed to humans. _Teams weighing how far to trust AI in support workflows can follow real deployment numbers like these on daily.dev._ ### What is Everest, Airbnb's internal AI context graph, used for? Everest is an internal organizational context graph using LLMs, embeddings, and AI-based retrieval that helped Airbnb launch grocery delivery and airport pickup services quickly. Learnings from building the grocery service, which took eight to nine months, were captured in Everest and let the airport pickup team finish its similar API-integration project in about six weeks, letting generalists work on specialist code. _Developers exploring how context graphs speed up cross-team projects can track approaches like this on daily.dev._ ## Similar posts on daily.dev - [Airbnb says AI now writes 60% of its new code](https://daily.dev/posts/airbnb-says-ai-now-writes-60-of-its-new-code-w24zdu6nr) · TechCrunch · 0 upvotes · 0 comments - [Airbnb’s CEO says nobody builds AI for normal people. He is on Y Combinator’s board](https://daily.dev/posts/airbnb-s-ceo-says-nobody-builds-ai-for-normal-people-he-is-on-y-combinator-s-board-cjw9v4ueq) · The Next Web · 0 upvotes · 0 comments - [Airbnb plans to bake in AI features for search, discovery and support](https://daily.dev/posts/airbnb-plans-to-bake-in-ai-features-for-search-discovery-and-support-2c40mwnl0) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience","url":"https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51"},"datePublished":"2026-10-02T14:11:02.772Z","dateModified":"2026-10-02T14:11:25.574Z","description":"Airbnb CTO Ahmad Al-Dahle, formerly head of generative AI at Meta, describes an 'inside-out AI' strategy: using AI internally to speed product development,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/d8797a82779e321a60d83ff2aaa2a31f?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/d8797a82779e321a60d83ff2aaa2a31f?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Latent Space","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Latent Space","logo":"https://media.daily.dev/image/upload/s--DnJ9laFj--/f_auto,q_auto/v1780213196/logos/latentspace?_a=BAMAMiWQ0","url":"https://daily.dev/sources/latentspace"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ai-agents","timeRequired":"PT8M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Latent Space","item":"https://daily.dev/sources/latentspace"},{"@type":"ListItem","position":3,"name":"Inside-Out AI: Rebuilding Airbnb Behind the Scenes and Across the Guest Experience"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/inside-out-ai-rebuilding-airbnb-behind-the-scenes-and-across-the-guest-experience-tnlc2rs51#faq","mainEntity":[{"@type":"Question","name":"What percentage of Airbnb's code is now written by AI?","acceptedAnswer":{"@type":"Answer","text":"About 60% of Airbnb's code is AI-authored. This shift accompanied roughly 80% more features and improvements shipped year over year, and a 1.6x increase in pull-request throughput per average engineer, driven by moving product, design, and engineering teams directly to working with prototypes instead of handoff-heavy processes. Engineering leaders tracking AI's real impact on software delivery follow case studies like this on daily.dev."}},{"@type":"Question","name":"How much of Airbnb's customer support is handled by AI agents?","acceptedAnswer":{"@type":"Answer","text":"Roughly half of Airbnb's support tickets are resolved purely by AI, a figure close to the nearly 45% cited in the company's Q2 2026 financial results. The system is built by generating a large battery of synthetic data to test the agent thoroughly before production, while certain sensitive cases like safety issues are deliberately routed to humans. Teams weighing how far to trust AI in support workflows can follow real deployment numbers like these on daily.dev."}},{"@type":"Question","name":"What is Everest, Airbnb's internal AI context graph, used for?","acceptedAnswer":{"@type":"Answer","text":"Everest is an internal organizational context graph using LLMs, embeddings, and AI-based retrieval that helped Airbnb launch grocery delivery and airport pickup services quickly. Learnings from building the grocery service, which took eight to nine months, were captured in Everest and let the airport pickup team finish its similar API-integration project in about six weeks, letting generalists work on specialist code. Developers exploring how context graphs speed up cross-team projects can track approaches like this on daily.dev."}}]} ```