Airbnb has a wide variety of ML models, and the ability to build, iterate, and maintain them is critical to its success. Many ML platforms cover data collection, feature engineering, training, deployment, and monitoring but few, if any, do all of the above seamlessly. Bighead aims to tie together various open-source and in-house projects to remove incidental complexity from ML workflows. This talk covers the architecture, the problems that each individual component and the overall system aim to solve, and a vision for the future of machine learning infrastructure. Bighead is widely adopted within Airbnb, already serving production traffic to improve search ranking, classify listing images, and detect fraudulent users. With the help of the framework, the model development time has been reduced from months to days. The project will be open sourced to allow wider community to benefit from its powerful features.
Senior Software Engineer
12:10PM - Day 2