Business Policy Experiments Using Fractional Factorial Designs
At DoorDash, we constantly strive to improve our experimentation processes by addressing four key dimensions, including velocity to increase how many experiments we can conduct, toil to minimize our launch and analysis efforts, rigor to ensure a sound experimental design and robustly efficient analyses, and efficiency to reduce costs associated with our experimentation efforts.
The DoorDash retail shopping experience mission seeks to combine the best parts of in-person shopping with the power of personalization.
It is amusing for a human being to write an article about artificial intelligence in a time when AI systems, powered by machine learning (ML), are generating their own blog posts.
At DoorDash, we generate supply and demand forecasts to proactively plan operations such as acquiring the right number of Dashers (delivery drivers) and adding extra pay when we anticipate low supply.
In real-world forecasting applications, it is a challenge to balance accuracy and speed.
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.
While building a feature store to handle the massive growth of our machine-learning (“ML”) platform, we learned that using a mix of different databases can yield significant gains in efficiency and operational simplicity.
Building an ML-powered delivery platform like DoorDash is a complex undertaking.
One challenge in running our platform is being able to accurately track Merchants’ operational status and ability to receive and fulfill 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?
