If you are among the tens of millions of customers who have shopped at Wayfair, you have experienced one of the many technologies built by the Machine Learning and Data Platforms team. We serve the business, engineering and data science teams across Wayfair, who are working on solving one of the most intriguing and gratifying challenges — helping our customers make their dream home a reality.
This large and growing team of engineers, product managers and designers is on the mission to build world class Machine Learning (ML) and Data technologies that enable us to make context-aware, real-time and intelligent decisions across every aspect of our business.
You can observe the value this team delivers to our customers throughout their journey with Wayfair, from product search, selection, checkout, fulfilment and customer service. Intelligent systems behind the scenes aim to improve overall customer experience and satisfaction. We let data drive our business decisions, and hence believe it should be accurate, timely, accessible and actionable to power our ML and analytics systems. The technologies powering the ML models are made possible by the platforms we build.
Meet the three parts of this amazing team.
Machine Learning Platforms — Scaling the ML capabilities that enable innovation
The Machine Learning Platforms team is focused on paving the path for any Wayfair team to leverage ML when serving our customers. This team is streamlining the capabilities needed for the end-to-end ML lifecycle from building, training, experimenting, deploying to operationalizing models at scale. It does so with purpose-built in-house solutions, and with best-in-class open source and vendor technologies, such as those offered by Google’s Vertex AI platform.
Machine learning is an empirical and iterative process, so the ability for our teams to be able to iterate multiple times over the model lifecycle with little to no friction is key to our innovation velocity. This team’s mission is to “make doing ML at Wayfair easy!” It is helping data science teams across Wayfair build and deploy models through a paved path, abstracting away the engineering complexities when delivering ML solutions.
ML Platforms team is composed of engineers and product managers working together to build the next generation of machine learning technologies within Wayfair. As you could imagine, such an endeavor would require putting together a multidisciplinary group. So we did just that. We brought together the top talent — from DevOps, Infrastructure, Machine Learning, Data Science, Software Engineering, and Analytics backgrounds to form this team.
Over the past few years this team has grown in size and so has its impact. Today we power several ML models across Wayfair. This team has delivered several state of the art capabilities for scaling model development and training, a feature library that serves both offline and online models, centralized model registry and a highly scalable model deployment platform.
We have a saying at Wayfair, “We are never done!” It is embodied in the mission of this team. While we are proud of our achievements to date, we have an intriguing and challenging product roadmap ahead of us that will take our platforms to the next level. We believe we are just getting started.
Data Engineering — Combining business & tech to fuel decisions at scale
At Wayfair, Data Engineering is a software engineering team with a specialization in building scalable and performant data platforms, data assets, and data pipelines that enable analysis, insights, and decisions. The downstream consumers of these platforms and assets are analytics, reporting, data science, business analysts, and other engineering teams; driving insights and decisions for the business, the customer, and the supplier. If you have heard the cliche metaphor of “Data is the new oil and analytics is the combustion engine.” Within that context, Data Engineering is all the components involved in getting the oil from the well to a gas pump.
Wayfair exemplifies a data-driven organization, where every decision and opportunity is instrumented on data. Data Engineering plays a critical role in architecting, developing, productizing, and automating trustable and timely data to drive business impact. Many of our business stakeholders have Python, SQL, and analytic skills comparable to developers anywhere else.
Organizationally, Data Engineering operates within the software engineering organization alongside Product Management, Machine Learning, and Data Science. This alignment provides us the close collaboration required to own end-to-end business domains of data, in the single-threaded ownership model, which Wayfair embraces. We bring strong collaboration across various teams, to enable the agility and speed that our hyper-growth organization needs.
Data engineers at Wayfair bring technical specialization within data engineering and expertise in business domains they partner with. Hence they need to have both technical chops and business acumen, providing them the ability to hold a strategic conversation about the business domain, its opportunities, and its future roadmap. Data Engineers bring vast expertise in architecting, modeling large-scale data warehouses, data lakes, and data platforms, building streaming, batch and micro-batch data pipelines, and governance and orchestration of data.
Customer Intelligence — Building platforms to enrich our world-class shopping experience
Customer Intelligence is composed of platforms aimed at empowering every Wayfair team with high quality data and the tools to activate it. It accomplishes this through ingesting, organizing and activating data in a scalable fashion. The platforms were born out of an original mission to produce insights that would power a single view of the customer for far-reaching purposes, from personalization to predicting behaviors to analytics. Today, those same tools have grown to ingest any type of data (product, customer, order, etc.) in real-time, support any type of prediction (e.g. shipping times, best email send time, product class affinity) and activate that data in all the ways teams desire to consume it.
- Scribe is a general purpose event definition and real-time tracking platform built by our technologists. As the core ingestion platform for the company – collecting upwards of 20 terabytes per day – Scribe enables applications all across Wayfair to react to events like add-to-carts and products going out of stock in real-time. Scribe also records its event data into long-term data stores for widespread use.
- Customer Identity Platform provides a comprehensive customer profile resolved across devices, accounts, and platforms to enable machine learning, personalization, customer service, and marketing at Wayfair. By providing access to a rich profile of customer attributes and insights, we enable Wayfair to provide a best-in-class experience and improve customer satisfaction, repeat rate, and lifetime-value. To deliver on our mission, we ingest data from a breadth of datasets across the business and transform those ingredients into digestible insights in the customer profile. These data points include recently viewed products, purchased classes, in-market score, B2B customer likelihood, first purchase date, and more. Our teams are able to segment customers based on profile attributes and understand the customer journey across a fragmented landscape of devices and channels, while prioritizing privacy and security. This platform is the backbone for all personalization activities onsite and in Owned and Operated channels like email.
Together, we empower Wayfair’s business and data science teams to make data-driven decisions at scale by providing data platforms and machine learning capabilities.
P.S. We are always looking for talented engineers and product managers, if our mission sounds interesting to you, please come talk to us!
Want to work at Wayfair? Head over to our Careers page to see our open roles. Be sure to follow us on Twitter, Instagram, Facebook and LinkedIn to get a look at life at Wayfair and see what it’s like to be part of our growing team.