Machine Learning - DoorDash Engineering Blog

Newest Article

Machine Learning Using Triplet Loss and Siamese Neural Networks to Train Catalog Item Embeddings

Learn how DoorDash used neural networks to better understand the contents of its large online catalog and improve search and recommendations

Abhi Ramachandran

19 Minute Read

Machine Learning

Increasing Operational Efficiency with Scalable Forecasting

Scaling forecasting to a large data team is not practical without a scalable platform. Learn how we built forecast factory at DoorDash

Ryan Schork 11 Minute Read

Machine Learning

Using ML and Optimization to Solve DoorDash’s Dispatch Problem

DoorDash delivers millions of orders every day with our last-mile logistics platform. Look under the hood and learn how the platform works

Alex Weinstein

Jianzhe Luo 18 Minute Read

Machine Learning

Predicting Marketing Performance from Early Attribution Indicators

Learn how DoorDash managed to respond faster to changes in the market by building a model to utilize early attribution data from experiments

Zhe Mai 10 Minute Read

Machine Learning

Managing Supply and Demand Balance Through Machine Learning

Achieving DoorDash's objectives requires a good balance between the supply of Dashers and the demand for orders. Learn how we manage this with ML

Stas Sajin

Zainab Danish 18 Minute Read

Data Machine Learning

Maintaining Machine Learning Model Accuracy Through Monitoring

Machine learning model drift occurs as data changes, but a robust monitoring system helps maintain integrity.

Swaroop Chitlur

Kornel Csernai 18 Minute Read

Machine Learning

Improving ETA Prediction Accuracy for Long-tail Events

Learn how we managed to better predict long tail delivery estimations using historical and realtime features as well as custom loss functions

Dawn Lu

Pratik Parekh 12 Minute Read

General Machine Learning

Best Practices for Regression-free Machine Learning Model Migrations

Migrating functionalities from a legacy system to a new service is a fairly common endeavor, but moving machine learning (ML) models is much more challenging. Successful migrations ensure that the new service works as well or better than its legacy version. But the complexity of ML models makes regressions more likely to happen and harder ...

Ying Chi 14 Minute Read

Data Machine Learning

Building Riviera: A Declarative Real-Time Feature Engineering Framework

In a business with fluid dynamics between customers, drivers, and merchants, real-time data helps make crucial decisions which grow our business and delights our customers. Machine learning (ML) models play a big role in improving the experience on our platform, but models can only be as powerful as their underlying features. As a result, building ...

Allen Wang

Kunal Shah 17 Minute Read

Machine Learning

Why Good Forecasts Treat Human Input as Part of the Model

At DoorDash, getting forecasting right is critical to the success of our logistics-driven business, but historical data alone isn’t enough to predict future demand. We need to ensure there are enough Dashers, our name for delivery drivers, in each market for timely order delivery. And even though it seems like people’s demand for food delivery ...

Brian Seo 10 Minute Read

Data Machine Learning

How to Drive Effective Data Science Communication with Cross-Functional Teams

Analytics teams focused on detecting meaningful business insights may overlook the need to effectively communicate those insights to their cross-functional partners who can use those recommendations to improve the business. Part of the DoorDash Analytics team’s success comes from its ability to communicate actionable insights to key stakeholders, not just identify and measure them. Many ...

James Williams

Lokesh Bisht 11 Minute Read

Data Machine Learning

Running Experiments with Google Adwords for Campaign Optimization

Running experiments on marketing channels involves many challenges, yet at DoorDash, we found a number of ways to optimize our marketing with rigorous testing on our digital ad platforms. While data scientists frequently run experiments, such as A/B tests, on new features, the methodology and results may not seem so clear when applied to digital ...