Airbnb’s Payments system needs to constantly evolve in order to serve both the growth of our business and complexity in operating a worldwide marketplace accurately and efficiently. There are dozens of payment method types, receiving & distributing many currencies with over 50 external vendor service integrations, handling billions in payment volume each year, all while ensuring that our marketplace operates in a safe and compliant manner with safeguards against ever evolving fraud.
Join us on Monday, October 03, at 12PM PT for a virtual tech talk with Mini Atwal (Head of Payments Engineering), and the Payments team, on Airbnb's next generation Payments architecture.
For most machine learning models, the more training data you provide, the better are the predictions. A common way to train a model is using a supervised learning technique. This requires humans to manually label data which is resource intensive and in some cases may not even be performant. In this talk, we will cover how Airbnb solves this problem by using self supervised learning, a method that trains models on raw unlabelled data, a more hands-off approach. Self supervised learning is a predictive learning, leveraging the underlying structure in the data to predict hidden parts. Join us, to learn more about how Airbnb uses this technique to solve problems that range from finding similar text to detecting fraud trends.
Airbnb’s expanding global presence requires its payments platform to add more capabilities every year. To take the next step in our architecture, we migrated our payments platform from a monolithic architecture to a service oriented architecture (SOA), separating our system to many bounded domains. This rehaul introduced significant additional complexity to data read flows as a side effect.
In this talk we’ll take a closer look at Airbnb Payments’ Unified Data Read Layer, which is designed to unify integration points for all payments data read clients under simple and extensible APIs, while hiding away internal payments system complexities and migrations. We’ll also give a few highlights from our onboarding so far, where we saw significantly faster load times.
Airbnb’s payment orchestration system is responsible for ensuring reliable money movement between hosts, guests, and Airbnb. As part of the migration to a service-oriented architecture (SOA), the Airbnb payments organization decided to fundamentally redesign this system to provide a sound technical foundation for future growth.
This talk will briefly explore why we chose to redesign the system, some key technical decisions, how we maintained reliable money movement during the migration, and key learnings from the experience.
Airbnb Payments enables both payins from Guests and payouts to hosts at a tremendous scale. As part of our global expansion, being able to rapidly integrate with new payment vendors is crucial to our success. In this talk, we will share how we design and implement a payment architecture that not only enables us to build new integrations easily, but also to maintain existing integrations efficiently.
Abhishek manages the Wallet Engineering teams at Airbnb. He also facilitates some of the overall Payments LTA efforts, promotes a multiplier mindset for EMs and is part of the Diversity and Belonging efforts of the company. Prior to Airbnb, he managed the Core Payroll domain at Zenefits for more than 2 years and preceding that he worked at PayPal/Venmo for 8 years building their retail payments platform.
Mini leads the technical teams for Payments, Risk and Compliance. With over 28 years of experience, Mini is a seasoned technology leader who has managed large global payments and credit teams for companies including PayPal and Funding Circle. Most recently, she served as SVP, Engineering at Live Nation/Ticketmaster, responsible for their B2C Marketplace.
Ali is working on Payments Business Integrations team as a senior full stack engineer. Currently he’s leading efforts on Airbnb Payments’ read optimized data store to materialize and serve denormalized payments data for simple and performant client integrations. Before joining Airbnb, he spent 5 years at Microsoft.
Xuemei Bao is working on the Payments Transactions team as a Tech Lead Manager. She is leading the efforts on building the next-generation billing orchestration system to ensure timely, reliable and safe money movement from guests and to hosts. She joined Airbnb in 2019 after spending 5 years at Facebook.
