Human feedback
Incorporate human insight to understand and improve quality
Capture domain expert feedback to understand how your app should behave and align your custom LLM-judge metrics with those expert's judgement.
Capture end user feedback to quickly pinpoint quality issues in production.
Intuitive labeling UIs for business users
MLflow's Review App enables busy domain experts to quickly provide feedback on production logs. Share logs for review and use predefined or custom questions.
Track and visualize feedback
MLflow replaces spreadsheets by attaching expert/user feedback to traces as versioned labels. Visualize this data in MLflow Trace UIs and dashboards to swiftly identify quality issues.
Capture end-user feedback
MLflow scalable feedback APIs allow you to attach end-user feedback from your deployed app to the source MLflow Trace, so you debug negative feedback with access to the step-by-step execution.
Integrated Chat App
Deploy new app versions to the Review App's chat UI. Domain experts can interact, give instant feedback, and help rapidly assess quality and pinpoint issues.
Get started with MLflow
Choose from two options depending on your needs
GET INVOLVED
Connect with the open source community
Join millions of MLflow users