Airbnb engineers detail how they adapted search ranking for map interfaces by modeling user attention through three iterations — uniform distribution, tiered markers, and decaying attention from map center — validated via A/B tests that increased bookings by 0.27% and reduced map interactions by 1.5%.
Airbnb rebuilt its authentication system using a server-driven 'identify first, challenge later' model, reducing client code by 60%, cutting bundle size by 100KB, enabling 20+ experiments in three months, and increasing successful logins by 2.6% while cutting duplicate accounts 27% and OTP costs 11%.
Airbnb built a Transformer‑based sequence model that encodes years of guest behavior—including long‑term bookings and short‑term browsing—to deliver timely, personalized search results, achieving up to 3.78% overall ranking improvement and significant gains in bookings and clicks.
Over five years, Airbnb's Viaduct service mesh grew traffic eightfold, doubled the number of owning teams to over 130, tripled its codebase to 1.5 million lines, halved incident minutes, and kept operational costs linear with QPS, leading to a major redesign and open‑source release.
Airbnb explains its three‑part Listing Lifetime Value framework—baseline, incremental, and marketing‑induced LTV—detailing the machine‑learning models, data collection, and the practical challenges of accurate estimation, incrementality, and uncertainty during market shocks.
This article explains how Airbnb broke a dangerous circular dependency in its observability stack by isolating metric collection onto dedicated Kubernetes clusters, adding a custom L7 network layer to decouple from the service mesh, and implementing meta‑monitoring with a dead‑man’s‑switch to keep monitoring systems reliable during failures.
Airbnb’s engineering team redesigned its payment platform to support over twenty local payment methods worldwide within fourteen months, using a domain‑driven, configuration‑driven architecture, standardized payment flows, multi‑step transaction handling, and a unified observability framework to boost conversion, expand markets, and improve reliability.
Airbnb transformed nearly 3,500 Enzyme test files to React Testing Library in just six weeks by building a large‑language‑model‑driven pipeline that validates, rewrites, retries, and enriches prompts with extensive context, achieving a 97% migration success rate while dramatically cutting manual effort and cost.
Airbnb completely rewrote its internal key‑value store Mussel, moving from a legacy V1 system to a cloud‑native, NewSQL‑backed V2 that reduces operational complexity, improves scalability, adds flexible consistency, and supports massive batch imports, all while achieving zero‑downtime migration of over 1 PB of data.
This article explains how Airbnb’s internal Impulse load‑testing‑as‑a‑service framework combines a load generator, dependency simulator, traffic collector, and test‑API generator to provide decentralized, containerized performance testing that integrates seamlessly with CI/CD pipelines and mimics production traffic.
Airbnb's media team engineered a scalable postcard generation pipeline that matches destinations, handles localized text layout, and delivers high‑performance images across iOS, Android, Web, and external messaging platforms using flexible templates, async processing, and pre‑generation strategies.
Airbnb’s Automation Platform v2 replaces the rigid, workflow‑driven architecture of v1 with an LLM‑centric design that orchestrates context gathering, chain‑of‑thought reasoning, tool execution, and guardrails, enabling more natural, scalable, and safe conversational AI while preserving the reliability of traditional workflows.
Airbnb tracks five real‑user performance metrics—Time To First Contentful Paint, Time To First Meaningful Paint, First Input Delay, Total Blocking Time, and Cumulative Layout Shift—using a mix of browser APIs and custom polyfills, combines them into a weighted Page Performance Score, and leverages that score to guide trade‑off decisions and detect regressions.
Airbnb created the Page Performance Score (PPS), a unified 0‑100 metric that aggregates platform‑specific initial‑load and post‑load user‑centric measurements for Web, iOS, and Android, using weighted curves to enable cross‑page, cross‑team comparisons, track organizational weighted averages, and evolve with new metrics while preserving a stable scale.
Airbnb’s Attribute Prioritization System uses a machine‑learning pipeline called LATEX to extract and map guest‑mentioned amenities, activities and places from reviews, messages and tickets, then predicts and ranks the most important attributes per listing, giving hosts personalized suggestions to improve listings and match traveler needs.
Airbnb’s Riverbed framework unifies streaming CDC events and batch Spark jobs behind a GraphQL‑based declarative API to automatically build and maintain distributed materialized views, using Kafka‑partitioned ordering and version control to deliver billions of daily updates with low‑latency reads for features such as payments and search.
Airbnb’s Listing Attribute Extraction Platform (LAEP) uses a custom NER model, word‑embedding mapping, and a BERT‑based scorer to automatically pull, normalize, and validate structured attributes from hosts’ unstructured text, boosting coverage for downstream tools and enhancing guest‑host matching at scale.
Airbnb’s new computer‑vision pipeline trains a deep‑learning aesthetic model with an EMD loss to rank photos, automatically sorts new‑listing images by design and room type, and scales real‑time similarity search via HNSW‑based ANN on AWS OpenSearch, boosting click‑through, bookings, and enabling unsupervised visual recommendations.
Airbnb’s three‑stage migration from its custom Epoxy/UIKit stack to SwiftUI involved extensive evaluation, rebuilding the design system with flexible style protocols, bridging UIKit and SwiftUI, adopting one‑way data flow, extensive testing and training, resulting in six‑fold code reduction, comparable performance, widespread production adoption, and higher developer productivity despite remaining framework challenges.
Airbnb upgraded its neural‑network ranking system by adding a similarity network that penalizes duplicate‑like listings, enabling the algorithm to present a more diverse set of options, which boosted booking rates, value, and five‑star ratings, demonstrating that reduced result similarity improves overall search quality.
Airbnb replaced its XLM‑RoBERTa ranking with an MT5 encoder‑decoder for content recommendation, built a real‑time generative assistant for reply suggestions and intent detection, and deployed a T5‑based paraphrase chatbot, showing that large‑scale pre‑trained transformers improve relevance, agent efficiency, and user satisfaction.
Airbnb’s Data Protection Service unifies privacy and security metadata, offering APIs that automate annotation verification, export and IDL validation, data‑subject‑rights orchestration, and secret‑leak detection, while assigning ownership, minimizing manual effort, and ensuring global, consistent compliance across the platform.
Airbnb’s engineering team describes how they migrated to Kubernetes, evolved their clusters through three stages, and built a custom gRPC‑based expander for the Cluster Autoscaler to achieve flexible, cost‑effective, and automated scaling across hundreds of heterogeneous clusters.
Airbnb’s Data Protection Platform (DPP) combines automated discovery, classification, encryption and privacy‑orchestration services—Inspekt, Angmar, Cipher, Obliviate, Minister, and the Madoka metadata system—to continuously inventory petabyte‑scale MySQL, Hive and S3 assets, track ownership and security attributes, and enforce GDPR, PIPL and CCPA compliance.
This article traces Airbnb's architectural journey from a single‑Ruby‑on‑Rails monolith through a microservices era to its current hybrid micro‑macro service model, highlighting the motivations, challenges, and engineering practices that shaped each transition.
Airbnb’s enterprise‑grade Automation Platform unifies conversational‑AI products through a Markov Decision Process‑based workflow system, featuring an Event Orchestrator, Workflow Engine, Action Store, and drag‑and‑drop Flow Builder, enabling rapid, scalable development of chatbots and assistants without deep engineering.
Airbnb’s Midas certification establishes a company‑wide, multi‑dimensional golden‑standard for data quality—covering accuracy, consistency, timeliness, cost, and completeness—by requiring collaborative design, automated health checks, and four review stages, ensuring certified data is reliable, well‑documented, and ready for reporting, experimentation, and machine‑learning.
Airbnb’s 2019 Data Quality Improvement Plan reorganized its data‑engineering workforce, introduced a dedicated data‑engineer role, adopted a decentralized Minerva‑based architecture with Spark pipelines, instituted rigorous testing, governance, and certification processes, and established SLAs and monitoring to ensure timely, trustworthy, well‑documented data across the enterprise.
This article explains the purpose and installation of EditorConfig, Prettier, and ESLint, demonstrates how to combine them with TypeScript, Airbnb style guide, and React Hooks, and provides a complete pre‑commit workflow using Husky and lint‑staged for consistent code quality.
This article reviews the Airbnb search ranking paper, detailing offline and online performance gains, the progression from SimpleNN to LambdaRankNN, GBDT/FM NN, and Deep NN models, failed embedding attempts, extensive feature engineering practices, and the production system architecture that enabled large‑scale deep learning deployment.
Nebula is a schema‑less, versioned data storage service built by Airbnb that unifies real‑time random access and offline batch processing, supporting low‑latency queries, incremental updates, and scalable snapshots using DynamoDB, HFileService, Spark pipelines, and Kafka streams.
This roundup highlights Google’s refreshed account management UI, Firefox’s renewed focus on privacy, and Airbnb’s decision to drop React Native in favor of native apps, outlining the technical and organizational challenges that drove the shift and its broader implications for mobile developers.
The article describes how Airbnb's data infrastructure team built a next‑generation log storage and query platform to improve data quality, timeliness, flexibility, and anomaly detection, outlining the system architecture, key requirements, five improvement areas, and the resulting benefits.
This article explores the cutting‑edge data pipelines of Netflix, Facebook, Airbnb and Pinterest, detailing the massive event volumes they handle, the core technologies such as Kafka, Spark, Presto and Hadoop, and how these giants design scalable, real‑time analytics infrastructures.
This article outlines the extensive technology stack behind Airbnb, covering its programming languages, client and server libraries, web server, data storage solutions, server services, and underlying hardware infrastructure that support its operations across 190 countries.
Airbnb’s engineering team outlines the evolution of its big‑data platform, detailing the philosophy behind its architecture, the dual “gold” and “silver” Hive clusters, migration to Mesos, use of Presto, Airpal, Airflow, and the performance and cost gains achieved through these design choices.
Airbnb’s engineering team outlines the evolution and design of its massive big‑data platform—detailing the dual “gold” and “silver” Hive clusters, use of Kafka, Presto, Airflow, Mesos, and Spark, along with performance gains, cost reductions, and open‑source contributions.
Airbnb tackled rapid data growth by vertically partitioning tables across multiple MySQL databases using Amazon RDS replication, minimizing downtime through read‑replica promotion, and implementing a staged migration that reduced primary database writes by 33%, cut size by 20%, and improved stability and scalability.
The article describes how Airbnb built and continuously improved a machine‑learning‑driven dynamic pricing tool—Aerosolve—that extracts property features, compares similar listings, incorporates seasonal and event‑driven demand, and automatically updates nightly price suggestions to help hosts set optimal rates.
Airbnb’s trust and safety team uses a series of machine‑learning models—starting from defining the prediction target, through data sampling and feature engineering, to evaluating precision and recall—to identify and mitigate fraud risks such as chargebacks across its global peer‑to‑peer rental platform.
This article chronicles how three financially strapped friends transformed a simple room‑rental idea into Airbnb, a $25.5 billion global platform, detailing their early challenges, Y Combinator acceleration, rapid growth strategies, and the broader impact on the sharing economy.
At Airbnb’s inaugural OpenAir conference, the company unveiled three open‑source big‑data tools—Airpal, a Presto‑based visual SQL query engine; Aerosolve, an interpretable machine‑learning engine for pricing recommendations; and Airflow, an internal platform for orchestrating and monitoring data pipelines.
The article recounts Riley Newman's five‑year journey at Airbnb, describing how the data science team evolved from a small, centralized group to an embedded, cross‑functional partner that treats user behavior as data, democratizes insights, and drives product and business decisions at scale.
