Databricks Apps vs Model Serving: Authentication, Cost, and Performance Compared
Summary
Databricks Apps are now the recommended first choice for deploying agents due to their flexibility in handling full-stack applications with multiple components, offering faster iteration and local testing compared to Model Serving. Model Serving remains suitable for use cases prioritizing high QPS, governance features like AI Gateway, inference tables, and guardrails, or when scaling to zero is acceptable for cost optimization.
Summary generated by brickster.ai from the video transcript.
More from Databricks Skill Builder
NewsDatabricks AI: PII Blocking for Data Privacy!
Databricks PII blocking creates input guardrails that prevent personally identifiable information from reaching AI models before processing. The feature allows users to configure and apply PII blocking policies across their account to protect sensitive data.
TutorialsHow to Connect Databricks and Fabric (No Copying Required)
Databricks catalog federation and Fabric mirroring enable zero-copy data access between platforms, allowing Fabric compute to read Databricks data or Databricks compute to read OneLake data without duplicating storage or managing ETL pipelines. The demonstration covers practical setup steps, required prerequisites like storage credentials and external data access permissions, and current limitations including read-only access and support for table objects only.
NewsPII Blocking in Action: Databricks AI Gateway Demo
Databricks AI Gateway's PII blocking feature detects and blocks personally identifiable information like emails, social security numbers, and addresses in user inputs before they reach external models. The demo shows that requests containing PII are rejected by the input policy, preventing sensitive data from being transmitted to the underlying model.
TutorialsPower BI to AI BI: The Migration Shortcut You Need
The video demonstrates how to use Databricks Genie Code to automate the migration of Power BI report files into Databricks AI BI dashboards and semantic models. It walks through preparing a Power BI file, launching the import tool, generating the dashboard, and enabling an integrated Genie space for natural language data querying.
TutorialsBuilding Agents on Databricks with Custom Apps and Omnigent
The video demonstrates how to build, update, and govern a store operations AI agent on Databricks using Model Context Protocol servers and custom apps. It shows how to use Omnigent and CodeX to add new context and tools, redeploy the application, and manage governance and traces through the Unity AI gateway.
TutorialsDon't Let AI Run Wild: Unity AI Gateway to the Rescue
The Unity AI Gateway centralizes the monitoring, management, and governance of enterprise AI models, coding agents, and MCP servers within Databricks. Administrators can use the platform to track usage metrics, configure spending budgets, set up external model providers, and enforce input policies that block sensitive data like PII.