Databricks Table Documentation

This repository contains comprehensive documentation for Narrative's Databricks warehouse tables, organized by product and functional area for easier navigation and query development.

Overview

Narrative uses Databricks as its primary data warehouse, containing tables across three main schemas:

• narrative.default - Primary operational tables
• narrative.core - Core business entities and logic
• narrative.stripe - Stripe integration for billing and payments

Repository Structure

.
├── databricks-docs/          # Organized table documentation
│   ├── products/            # Product-specific tables
│   ├── business/            # Business logic and operations
│   ├── analytics/           # Event tracking and analytics
│   └── infrastructure/      # System and ML operations
├── query_context.md         # Original monolithic documentation (3.3MB)
├── CLAUDE.md               # Instructions for Claude AI assistant
└── README.md               # This file

Quick Start

Finding Tables by Product

• Narrative Select (AI culling app): databricks-docs/products/select_tables.md
• Narrative Publish (blogging tool): databricks-docs/products/publish_tables.md
• Edit Features: databricks-docs/products/edit_tables.md

Finding Tables by Function

• User Management: databricks-docs/business/users_accounts.md
• Billing & Payments: databricks-docs/business/billing_stripe.md
• Event Tracking: databricks-docs/analytics/events_tracking.md
• ML/AI Systems: databricks-docs/infrastructure/ml_inference.md

Table Statistics

Category	File Count	Table Count	Key Areas
Products	3	66	Select (47), Publish (4), Edit (15)
Business	3	97	Stripe (71), Core (20), Users (6)
Analytics	2	9	Events (8), Web (1)
Infrastructure	3	24	Assets (5), ML (3), Operations (16)
Total	11	196

Usage Examples

For Product Managers

-- Find Select activation metrics
-- See: databricks-docs/products/select_tables.md
SELECT * FROM narrative.default.select_activation
WHERE created_at >= CURRENT_DATE - INTERVAL 30 DAYS;

For Data Analysts

-- Analyze user engagement across products
-- See: databricks-docs/analytics/events_tracking.md
SELECT event_type, COUNT(*) as event_count
FROM narrative.default.select_events
GROUP BY event_type
ORDER BY event_count DESC;

For Engineers

-- Check ML inference performance
-- See: databricks-docs/infrastructure/ml_inference.md
SELECT * FROM narrative.default.cloud_xmp_inference_params_staging
WHERE sync_status = 'pending';

Navigation Guide

Detailed navigation and usage instructions are available in databricks-docs/README.md.

About Narrative

Narrative is an Auckland-based startup (founded 2017) building AI-powered desktop software that helps photographers streamline their post-shoot workflow. Our main products are:

• Narrative Select - AI culling app for rapid photo selection
• Narrative Publish - Fast blogging tool for photographers

Contributing

This documentation is auto-generated from our Databricks warehouse. To update:

1. Re-export table documentation from Databricks
2. Run the splitting script to reorganize into categories
3. Submit a pull request with the changes

License

This documentation is for internal use by Narrative team members and authorized partners.