Autoplay
Autocomplete
Previous Lesson
Complete and Continue
Databricks Data Quality Monitoring with Zapier
Introduction
1.1 The Three Things You'll Build: A Chat, A Dashboard, An Alerter (5:00)
1.2 Our Problem: A Failed Check Doesn't Tell You Severity, Owner, or Next Step (5:29)
1.3 How MCP Actually Works: Zapier Serves the Menu, Claude Does the Ordering (2:28)
Observe and Explain, the Zapier MCP Server
2.1 Load the Run Results into Databricks and the Runbook into Google Sheets (5:51)
2.2 Build One MCP Server for Both Sources and Test It Before Any App Code (11:23)
2.3 Build the Streamlit Chat and Let Claude Run the Tool Calls (9:16)
2.4 Add ChatGPT as a Zapier Action and Have Claude Call It (6:45)
Detect and Act, the Zapier SDK
3.1 Explore Your Connections with the CLI, Then Set Up the SDK with Claude Code (5:24)
3.2 Build the Dashboard That Pulls Your Run Metrics Through the SDK (7:30)
3.3 Build the Quality Alerter: Join Each Failure to Its Owner (5:55)
3.4 Send a Real Gmail Alert to the Owner with Remediation Steps (4:33)
Summary and Outlook
4.1 Recap: When to Use MCP, When to Write Code, Where to Go Next (3:51)
2.1 Load the Run Results into Databricks and the Runbook into Google Sheets
Lesson content locked
If you're already enrolled,
you'll need to login
.
Enroll in Course to Unlock