When the Yelp Android App was first built it was a monolith, with almost everything inside a single module. Single modules are easier to deal with; there are no complex dependency graphs to account for and some details of how the app is managed, such as how testing is implemented, are simpler. Eventually the app and company structure grew to a size where the monolith made code changes and testing harder, where modularization would help the app scale. After some discussion we decided to break the app into modules, thinking carefully about how we structured the app. We’ll discuss our...

Continue reading

Since its inception in 2013, Yelp has grown its transactions platform to tens of millions of orders. With this growth, it’s become slow and cumbersome to rely solely on MySQL for searching and retrieving user orders. Transactions at Yelp are built around integrations with dozens of external fulfillment partners across different verticals, from food ordering to spa reservations. In most of these integrations, Yelp takes care of payment processing. Given this, the way we store order details prioritizes facilitating stateful order processing in a uniform way. However, other access patterns like finding popular orders by category or fetching a user’s...

Continue reading

In our previous blog post, we established a performance improvement lifecycle and explored the first step by defining what metrics we wanted to measure and establishing a baseline on which we can improve. In this blog post, we’ll cover how we improved our two metrics of initial render timings and scroll performance when rendering search results in the Yelp Android app. Step 2: Making Performance Improvements Not all changes we made were specific to only one of our goals. Often times, a change that was expected to contribute to one goal also had an effect on the other. We used...

Continue reading

In our previous post, we discussed how we used PySpark to build a large-scale distributed machine learning model. In this post, we will describe our experience and some of the lessons learned while deploying PySpark code in a production environment. Yelp’s systems have robust testing in place. It’s a hallmark of our engineering. It allows us to push code confidently and forces engineers to design code that is testable and modular. Broadly speaking, we found the resources for working with PySpark in a large development environment and efficiently testing PySpark code to be a little sparse. By design, a lot...

Continue reading

Here at Yelp our core mission is to connect users with great local businesses. On the engineering side, that requires us to tackle big data and machine learning problems. One of the natural ways to apply machine learning to our core mission is to deliver relevant and personalized business recommendations. This helps us deliver great content to users. For example, we can let users know when a hot and new coffee shop we think they will love opens in their neighborhood through push notifications or features like collections. This is an area of active research and development at Yelp, and...

Continue reading

#perfmatters At Yelp, we’ve been working hard to improve the performance of our search results on mobile devices (iOS and Android). We know that performance matters from a numbers perspective, but another reason why we’ve decided to invest more in performance is due to a recent increase in user studies commissioned by Yelp. We noticed that most users “grunted or made noises when waiting for search to load”. We took this as a sign that something needed to be done! In our quest to make search faster on the Android mobile app client, we broke down performance into two different...

Continue reading

What is Salsa? Salsa is an open source library that renders iOS views and exports them into a Sketch file. We built Salsa to help bridge the gap between design and engineering in an effort to create a single source of truth for visual styling of UI. Why use Salsa A few years ago, we started putting together a library of common components that developers and designers could use to build features. Initially, we had to manually maintain consistency between the designs in Sketch and their implementations in code. When we only had a handful components, this wasn’t difficult to...

Continue reading

Many enterprise professionals use passwords that are weak and easily compromised. Equipped with this knowledge, as well as the exposure of more and more password leaks, dictionary attacks focused on compromised or popular passwords have become increasingly effective. As such, the National Institute of Standards and Technology recommends password blacklisting as a highly-effective means of preventing such attacks. Unfortunately, use of password blacklisting countermeasures has remained a relatively new innovation that has not yet achieved widespread corporate adoption. At Yelp, however, we strive to add the latest and greatest defense mechanisms to our arsenal, which is why we adopted such...

Continue reading