You are responsible for:

• Working with internal customers (e.g. other teams inside the Wikimedia Foundation) and external customers (e.g. Wikipedia editors and other volunteers) to build, deploy, and manage productionized, scaled machine learning models. This includes communicating with the customers early to assess needs, working with them to scope out the appropriate tooling that might be needed, gathering the training data, training the model in a repeatable process, deploying the model on a deployment cluster, and monitoring that model over time.
• Helping other teams at the Foundation and the broader community understand the work conducted on the team.

Required skills and experience:

• 5+ years of experience in an MLE and/or MLOps role as part of a team maintaining production models
• Experience with popular Python ML libraries
• Strong ML background, particularly solving NLP tasks
• Strong English language skills
• Experience with global coworkers
• Experience with remote work and/or async work

Additionally, we’d love it if you have:

• Experience in fine-tuning pre-trained models
• Familiarity with ETL processes and data pipeline orchestration
• Experience with any pipeline orchestration tool (Airflow, Kubeflow, Argo workflows etc)