Join Our Team

Machine Learning

Machine Learning

4 Essential Steps for Building a Simulator

For complex systems simulating the impact of algorithmic changes is often faster and less costly than experimenting on features in production

Devjit Chakravarti
Machine Learning

Leveraging Causal Inference to Generate Accurate Forecasts

Learn how DoorDash captures hard to measure macroeconomic effects like IRS refunds and the effect of Daylight savings in these case studies

Chad Akkoyun Qiyun Pan
Data Machine Learning

Meet Dash-AB — The Statistics Engine of Experimentation at DoorDash

Learn how DoorDash was able to test uniformly according to established best practices and reuse complex statistical methods with Dash AB

Caixia Huang Yixin Tang
Machine Learning

Building the Model Behind DoorDash’s Expansive Merchant Selection

Having a quality selection did not happen by accident. Learn about the ML models that power the diverse, high quality selection on our platform

Lu Wang Ying Yang Chen Dong
Machine Learning

Using Gamma Distribution to Improve Long-Tail Event Predictions

Which loss function is best for long tail event prediction? learn how we used gamma distributions to boost eta prediction accuracy

Pratik Parekh Zhe Jia
Machine Learning

Using a Multi-Armed Bandit with Thompson Sampling to Identify Responsive Dashers

Epsilon-Greedy, The Upper Confidence Bound, Thompson Sampling: which is the best Multi-armed bandit algorithm for promotion optimization?

Arjun Sharma
Machine Learning

Ship to Production, Darkly: Moving Fast, Staying Safe with ML Deployments

Learn how DoorDash balanced ML models' release speed and reliability by shipping darkly in order to manage fraud model deployments

Bob Nugman
Machine Learning

Balancing Network Effects, Learning Effects, and Power in Experiments

Experimenting can mean balancing learning and network effects, while ensuring adequate power. Learn how we manage this trilemna in this post

Wei Feng Jared Bauman
Data Machine Learning

Introducing Fabricator: A Declarative Feature Engineering Framework

Learn how our Fabricator infrastructure integrations for ML features were automated and continuously deployed to generate 500 unique features and 100+ billion daily feature values

Kunal Shah
Machine Learning

How DoorDash Quickly Spins Up Multiple Image Recognition Use Cases

Learn how DoorDash utilizes deep learning to power some of its image recognition use-cases

Chi Zhang Sushil Vellanki
Machine Learning

Improving Subgroup Analysis with Stein Shrinkage

Learn how DoorDash Utilized Stein Shrinkage to perform subgroup analysis without the danger of high variances

David Kastelman
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

Subscribe to our Eng blog for updates

Thank you for subscribing!

Sign up for updates

Want More
Engineering Updates?

Susbscribe to the DoorDash engineering blog