Domain experts use this MLflow website to label traces or chat with an agent. Agents hosted on Databricks Apps lack the Chat UI and use labeling sessions for review instead.
Read more: Databricks docs
In the Ultra Transcenders books
Each book explains Review App in context, with comparison tables and the common traps.
Terms in this definition
- Domain
A way of grouping workspaces by area of the business, in support of a data mesh approach. Items take on their workspace's domain, letting you filter the OneLake catalog by it, and certain tenant settings can be passed to domain admins; domains have no effect on access permissions.
- MLflow
An open-source framework for experiment tracking and model management. Azure Machine Learning adopts it as its tracking API, so each workspace acts as a compatible MLflow tracking server.
- Route table label
Name for a set of route tables in Virtual WAN hubs, used when choosing where routes propagate; Default, for instance, covers the defaultRouteTable of every hub.
- Agent
A specialised form of Microsoft Copilot set up for one particular job, pairing instructions with knowledge and skills. You can create one in Copilot Studio, SharePoint or Agent Builder, and administrators control them from the Microsoft 365 admin center.
- Databricks Apps
A serverless service for hosting secure data and AI applications built with Python or Node.js. Unity Catalog governs them, and compute is charged by the hour while an app runs.
Related terms
- Retriever
The piece that looks up the chunks best matching a query, and can be exposed to an agent as a tool. Retrieval judges and the Review App locate its context through the
RETRIEVERspan it gets in MLflow tracing. - SME
Someone with the domain knowledge to judge whether an application answers correctly. They respond through the MLflow Review App, and what they label can serve as expectations in evaluation.