Application Development Questions

A production RAG application on a standard Mosaic AI Vector Search endpoint experiences increased query latency. The index holds 50 million vectors at...

When logging a LangChain or PyFunc AI agent with `mlflow.langchain.log_model()`, what does the `resources` parameter declare?

Which three task types are supported by the Foundation Model APIs `task` parameter when configuring a model serving endpoint for external models?

A RAG system returns 5 retrieved chunks per query. Evaluations show the system frequently includes irrelevant chunks when queries use nuanced paraphra...

A company deploys an LLM chatbot with variable, unpredictable traffic patterns including long idle periods. Which Foundation Model API billing mode is...

What are the two resources a developer must create before querying documents using Mosaic AI Vector Search?

Which MLflow function is used to log a LangChain agent using the code-based logging approach on Databricks?

A RAG system retrieves product support documents containing specialized component identifiers such as "Part-7X-A4-ENG22". Users often query using exac...

A developer queries a Mosaic AI Vector Search index to find relevant customer support documents for a question. They want to restrict results to docum...

An ML engineer logs a PyFunc agent that requires a Vector Search index and a Model Serving endpoint at inference time. The following code is used: ```...

A developer builds a PyFunc AI agent and wants to define the expected input and output schema for the serving endpoint to enforce validation. Which ML...

A team logs a LangChain LCEL chain using `mlflow.langchain.log_model(lcel_chain, artifact_path="agent")` and wants to run batch inference using the st...

Which MLflow Python package flavor is used to log and load PyFunc models that implement the `mlflow.pyfunc.PythonModel` interface?

A team uses LangChain to build a RAG chain. They want to add MLflow auto-tracing to automatically capture spans for each LangChain component (retrieve...

An engineer uses `mlflow.langchain.log_model()` to log a LangChain RAG chain. They want to ensure the chain's `pip_requirements` are correctly capture...