Learn how DoorDash optimizes supply and demand forecasts using a cascade machine learning approach for enhanced accuracy during holidays. Discover how this innovative technique, combining Gradient Boosting and linear models, ensures operational efficiency, top-notch customer experiences, and seamless Dasher interactions.
In real-world forecasting applications, it is a challenge to balance accuracy and speed. We can achieve high accuracy by running numerous models and configuration combinations and we gain speed through running fast, computationally inexpensive models. We explore a number of models and configuration combinations at DoorDash to forecast demand on our platform. However, the challenge ...
In the wake of ChatGPT and Generative AI DoorDash is identifying ways this new technology can enhance the customer’s ordering experience on the platform. The company is exploring the use of Generative AI, a subset of Artificial Intelligence that generates novel content based on existing data, and how it can be implemented effectively with consideration ...
Read how DoorDash's product development life cycle works in this new ML blog post about how we optimized when orders are sent to merchants
Learn how DoorDash engineering utilized ML models to o accurately track Merchants' operational status and ability to full fill orders.
In order to inspire DoorDash consumers to order from the platform there are few tools more powerful than a compelling image, which raises the questions: what is the best image to show each customer, and how can we build a model to determine that programmatically using each merchant’s available images? Out of all the different ...
When we encountered a business problem that required an error rate of zero, we turned to a classification model with human review
Learn how our model balances exploitation and exploration during ranking to optimize the consumer experience while simultaneously improving fairness for merchants
Data preparation, represents The vast majority of work in developing machine learning models, learn how to make things easier
When expanding from made-to-order food delivery to new product verticals like groceries, convenience, and retail, new challenges arise, including how to ensure inventory will be available to fulfill orders. As a business, we always want customers to receive all the items they ordered. For restaurant orders, this is easy to do because merchants offer relatively ...
For complex systems simulating the impact of algorithmic changes is often faster and less costly than experimenting on features in production
Learn how DoorDash captures hard to measure macroeconomic effects like IRS refunds and the effect of Daylight savings in these case studies
Susbscribe to the DoorDash engineering blog