model-serving

Deploy and query MLflow models and AI agents on Databricks Model Serving.

Updated Mar 10, 2026

Search Tags:#deployment #mlflow #databricks #api endpoints #model serving #genai agents

Ready to run

npx skills add https://github.com/slysik/databricks-claude-coding --skill model-serving-slysik

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: model-serving
Source: https://github.com/slysik/databricks-claude-coding/tree/main/.agents/skills/model-serving
Command: npx skills add https://github.com/slysik/databricks-claude-coding --skill model-serving-slysik

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides end-to-end guidance for deploying classical ML models, custom PyFunc models, and GenAI agents to Databricks Model Serving endpoints, streamlining the path from development to production.

Core Features & Use Cases

Model Deployment: Deploy various MLflow model types, including traditional ML, custom PyFunc, and GenAI agents.
Agent Development: Build and deploy conversational AI agents using ResponsesAgent and LangGraph, integrating tools like UC Functions and Vector Search.
Use Case: Deploy a scikit-learn model for customer churn prediction or deploy a GenAI agent that answers customer support queries using a foundation model and retrieves information from a vector database.

Quick Start

Use the model-serving skill to deploy the 'main.agents.my_agent' model version 1.

Frequently Asked Questions about model-serving

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I deploy an MLflow model to Databricks Model Serving?▼

Yes, you can deploy GenAI agents using Databricks Model Serving. This Skill supports building and deploying conversational AI agents using ResponsesAgent and LangGraph, including integration with UC Functions and Vector Search tools.

Does Databricks Model Serving support asynchronous deployment for custom PyFunc models?▼

Yes, Databricks Model Serving supports asynchronous deployment for custom PyFunc models. This Skill covers asynchronous deployment and package management to streamline the path from development to production.

What's the best way to integrate tools like Vector Search when serving AI agents?▼

The best way to integrate tools like Vector Search when serving AI agents is to build your agent using LangGraph or ResponsesAgent. This Skill facilitates integrating these tools during agent development and deployment.

Can I manage package dependencies when deploying custom PyFunc models to API endpoints?▼

Yes, you can manage package dependencies when deploying custom PyFunc models to API endpoints. This Skill provides guidance on package management to ensure your models run correctly in the production serving environment.

How do I query a deployed model version on a Databricks endpoint?▼

To query a deployed model version on a Databricks endpoint, you use the registered model version name within your API call. This Skill supports the full lifecycle including the querying of deployed MLflow models and AI agents.