Wayfair Tech Blog

A picture of a yellow couch with accent pillows and a coffee table can be used to extract product tags, like "square arm", "sofa", "vintage modern" style, etc. for the couch, and "table", "round", "black", "4 legs" for the coffee table.

Accelerating Catalog Tagging Automation with Snorkel’s Data-Centric AI Platform: Wayfair’s Success Story

One of our missions at Wayfair is to help our 22 million customers find the products they are looking for. For example, when a customer searches for a “modern yellow sofa” on Wayfair, we want to show the most relevant options from the tens of thousands of sofas available in our catalog. To do that, it is important to have a strong understanding of the products we sell. We use machine learning algorithms to analyze and understand the descriptive information (e.g. color, shape) of over 40 million products provided by more than 20 thousand suppliers in our catalog.

A graph network connects 3 account holders as well as two "synthetic persons" via credit cards, phone numbers, bank accounts, and social security numbers

Wayfair Tech Blog

Preventing Policy Abuse with Graph Neural Networks

How Wayfair leverages Graph Neural Networks to thwart account-hopping fraudsters

Wayfair Tech Blog

An Inside Look at the State of Retail at the Milken Institute Global Conference

Last week I participated on a panel at the Milken Institute Global Conference. It was a fantastic event featuring more than 200 sessions, including mine, “Retail Renaissance: Businesses Embracing Technology, Sustainability, and Innovation.”

Wayfair Tech Blog

Introducing Wayfair’s Digital Design Studio

In early April, Wayfair introduced the Digital Design Studio in our AllModern Dedham store. The Digital Design Studio (DDS) is a photorealistic interactive experience that lets customers explore Wayfair’s endless catalog by designing 3D furniture arrangements with product cards. Creating this tool was a collaborative effort between Wayfair Next (our R&D group tasked with exploring various emerging tech use cases), Physical Retail, Visual Merchandising, Creative, 3D, and many other Wayfair teams.

Left image shows population of millions of users being split into test groups A and B. Right images shows 50 states comprised of 210 geos being split int tests groups A and B

Wayfair Tech Blog

How Wayfair Uses Geo Experiments to Measure Incrementality

Geo experimentation unlocks incrementality testing when customer-level experimentation is not possible.

Wayfair Tech Blog

Delivery-Date Prediction

How to make deliveries faster without touching any packages

a flow diagram shows the transition between multiple wayfair webpages being represented by a vector of integers in an embedding space

Wayfair Tech Blog

Introducing Melange: A Customer Journey Embedding System for Improving Fraud and Policy Abuse Detection

Let’s look at how we at Wayfair use custom embedding models to capture complex customer behavior patterns for advanced Fraud prevention.

Wayfair Tech Blog

Wayfair releases WANDS, the largest and richest publicly available dataset for e-commerce product search relevance

Search relevance – the relationship between users’ queries and the products returned in search results – is one of the most important performance indicators for ecommerce storefronts. However, the sheer volume of the data makes evaluating and improving search relevance a difficult proposition. To give just one example, the Wayfair catalog has millions of products, which makes it difficult to create relevance-focused evaluation datasets manually.

CI/CD and Scheduling pipeline for Vertex AI

Wayfair Tech Blog

How Wayfair is reaching MLOps excellence with Vertex AI

Dive into the ways Wayfair’s data scientists are using Vertex AI to improve model productionization, serving, and operational readiness velocity.