Purpose: Comprehensive reference documentation for building framework wrappers around the Dataiku Python API.

Audience:

• Developers building frameworks/wrappers
• Claude Code sessions doing deep API work
• Anyone needing complete API reference

Sequential guides for understanding the entire API:

1. 01-prerequisites-and-setup.md

   • Project structure recommendations
   • API key setup and security
   • File naming conventions
   • Configuration patterns

2. 02-api-overview.md

   • API architecture and design
   • Main modules and their purposes
   • Package organization
   • Entry points and client types

3. 03-client-authentication.md

   • Authentication methods
   • Client initialization patterns
   • Connection management
   • Multi-instance scenarios

4. 04-project-operations.md

   • Complete project API reference
   • All project-level operations
   • Project settings and metadata
   • Project contents access

5. 05-dataset-operations.md

   • Dataset types and creation
   • Schema management
   • Data access patterns
   • Build and partition operations

6. 06-recipe-operations.md

   • Recipe types reference
   • Recipe creation patterns
   • Execution and monitoring
   • Schema propagation

7. 07-scenario-operations.md

   • Scenario structure
   • Steps and triggers
   • Execution patterns
   • Monitoring and logging

8. 08-job-operations.md

   • Job lifecycle
   • Status monitoring
   • Log access
   • Async patterns with futures

9. 09-ml-operations.md

   • ML task reference
   • Model training and evaluation
   • Saved model management
   • Deployment patterns

10. 10-admin-operations.md

    • User and group management
    • Connection management
    • Code environments
    • Instance settings

Deep technical reference:

• classes_and_methods.md - Complete class/method inventory (1,162 lines)
• class_index.md - Quick reference organized by category (500 lines)
• common_patterns.md - Workflow patterns and examples
• coverage_analysis.md - Gap analysis and recommendations
• README.md - How to use the inventory

Build a framework wrapper:

1. Start with documentation/01-prerequisites-and-setup.md
2. Read documentation/02-api-overview.md for architecture
3. Use api_inventory/classes_and_methods.md as complete reference
4. Check api_inventory/common_patterns.md for implementation patterns

Find a specific class/method:

1. Use api_inventory/class_index.md for quick lookup
2. Check api_inventory/classes_and_methods.md for details

Understand workflows:

1. Read relevant documentation/0X-*.md guide
2. Check api_inventory/common_patterns.md for code examples

Assess what's possible:

1. Skim api_inventory/class_index.md for categories
2. Review api_inventory/coverage_analysis.md for completeness

Use both together:

• claude-guides/ - For building Dataiku projects quickly
• dataiku_framework_reference/ - For building wrappers/frameworks

API Inventory:

• 150+ classes documented
• 1,000+ methods cataloged
• 7 client types covered
• 15 functional categories
• 50+ workflow examples

Reference Guides:

• 10 comprehensive sections
• Complete method signatures
• Parameter documentation
• Return type specifications
• Example code throughout

When building with this reference:

your_project/
├── config/
│   ├── APIKEY.txt                    # Your API key (never commit!)
│   └── config.json                   # Environment config
├── dataiku_framework_reference/      # This documentation (read-only)
│   ├── documentation/
│   ├── api_inventory/
│   └── README.md
├── claude-guides/                    # Usage guides (read-only)
├── your_framework/                   # Your wrapper code
│   ├── __init__.py
│   ├── client.py
│   ├── projects.py
│   └── ...
├── tests/                            # Your tests
├── examples/                         # Usage examples
└── {PROJECT_KEY}_script.py          # Project-specific scripts

1. Read prerequisites:

   cat documentation/01-prerequisites-and-setup.md

2. Understand API architecture:

   cat documentation/02-api-overview.md

3. Explore what's available:

   cat api_inventory/class_index.md

4. Dive into specific areas:

   cat documentation/04-project-operations.md  # For project wrappers
   cat documentation/05-dataset-operations.md  # For dataset wrappers
   # etc.

1. Find a class:

   grep "class DSSDataset" api_inventory/classes_and_methods.md -A 20

2. Find a method:

   grep "def build" api_inventory/classes_and_methods.md

3. Find patterns:

   cat api_inventory/common_patterns.md | grep -A 10 "Dataset Processing"

When updating:

1. Keep api_inventory/ as pure technical reference
2. Put workflow guidance in documentation/
3. Include method signatures and return types
4. Add examples for complex patterns
5. Update coverage analysis when finding gaps

• API Version: 14.1.3+
• Last Updated: 2025-11-21
• Python: 3.7+
• Total Lines: ~10,000+

• Usage Guides: ../claude-guides/ (for building projects)
• Official Docs: https://doc.dataiku.com/
• API Reference: https://developer.dataiku.com/latest/api-reference/python/
• GitHub: https://github.com/dataiku/dataiku-api-client-python

Ready to build? Start with documentation/01-prerequisites-and-setup.md