For any data-driven company, it’s key that every change is tested by experiments to ensure that it has a positive measurable impact on the key performance metrics.
Category Archives: AI & ML
Building the Model Behind DoorDash’s Expansive Merchant Selection
An optimized merchant selection strategy has been one of the key factors that has enabled DoorDash to become an industry leader in US food delivery service.
Using Gamma Distribution to Improve Long-Tail Event Predictions
For DoorDash, being able to predict long-tail events related to delivery times is critical to ensuring consumers’ orders arrive when expected.
Using a Multi-Armed Bandit with Thompson Sampling to Identify Responsive Dashers
Maintaining Dasher supply to meet consumer demand is one of the most important problems for DoorDash to resolve in order to offer timely deliveries.
Ship to Production, Darkly: Moving Fast, Staying Safe with ML Deployments
At DoorDash, machine learning (ML) models are invoked many millions of times each day.
Balancing Network Effects, Learning Effects, and Power in Experiments
At DoorDash, we rely on experimentation to make decisions regarding model improvements and product changes because we cannot perfectly predict the results in advance.
Introducing Fabricator: A Declarative Feature Engineering Framework
As machine learning (ML) becomes increasingly important across tech companies, feature engineering becomes a bigger focus for improving the predictive power of models.
How DoorDash Quickly Spins Up Multiple Image Recognition Use Cases
DoorDash has rich image data collected by Dashers, our delivery drivers, that we use in a number of use cases.
Improving Subgroup Analysis with Stein Shrinkage
DoorDash is often interested in knowing not only the average effect of an intervention, but also the more granular effect of that intervention on specific cohorts.
Using Triplet Loss and Siamese Neural Networks to Train Catalog Item Embeddings
Understanding the contents of a large digital catalog is a significant challenge for online businesses, but this challenge can be addressed using self-supervised neural network models.