Title of session: ML/AI models on LiftWing
Session description: Demo session about the different machine-learning models that can be accessed publicly via LiftWing, WMF's scalable machine learning model serving infrastructure on Kubernetes using KServe. The Research Team and the Machine Learning Team at WMF have developed and deployed a set of new models to automatically score the risk of a revert of an edit (to help editors identify edits that should be patrolled) or assess the topic or readability of articles (see reference). In this session, I am happy to talk about how these models work (e.g. how they were trained) and how they can be used in practice (e.g. how to query the model via the API); as well as what are our plans for the future. If there's anything specific you'd like to see demo'd or discuss, comment in the task.
Username for contact: @MGerlach
Session duration (25 or 50 min): ? (either is fine)
Session type (presentation, workshop, discussion, etc.): presentation/demo
Language of session (English, Arabic, etc.): English
Prerequisites (some Python, etc.): This session is targeted at folks who are interested in using ML models that are deployed on LiftWing and publicly accessible via the LiftWing-API for their own tools , but anyone is welcome.
Any other details to share?:
Interested? Add your username below:

• @roti_WMDE