ExOutlines Documentation Index

Complete index of all documentation, guides, examples, and tutorials.

Last Updated: 2026-01-28

Version: 0.2.0

Quick Links

• Getting Started - Start here if you're new
• API Reference - Complete API documentation
• Examples Directory - Production-ready code examples (see below)
• Livebook Tutorials - Interactive learning

Core Documentation

Guides (11 total)

Foundation (3)

1. Getting Started (450+ lines)

   • Installation and setup
   • First schema creation
   • Validation basics
   • Backend configuration
   • Common patterns

2. Core Concepts (800+ lines)

   • Structured generation principles
   • Schema definition
   • Type system
   • Validation process
   • Retry-repair loop
   • Backend architecture
   • Batch processing
   • Telemetry

3. Architecture (900+ lines)

   • System overview
   • Module organization
   • Validation engine
   • Backend implementations
   • Design decisions
   • Extension points

Schema Design (1)

4. Schema Patterns (600+ lines)
   • Basic patterns
   • String validation (length, regex, format)
   • Numeric constraints
   • Array patterns
   • Nested objects
   • Union types
   • Enum patterns
   • Common scenarios
   • Schema composition

Integration (2)

5. Phoenix Integration (500+ lines)

   • Controller integration
   • LiveView patterns
   • Background jobs with Oban
   • Caching strategies
   • Production deployment

6. Ecto Schema Adapter (400+ lines)

   • Converting Ecto schemas
   • Automatic validation extraction
   • Embedded schemas
   • Integration patterns

Testing and Quality (2)

7. Testing Strategies (550+ lines)

   • Unit testing with Mock
   • Integration testing
   • Property-based testing
   • CI/CD integration

8. Error Handling (500+ lines)

   • Diagnostics structure
   • Retry strategies
   • Graceful degradation
   • Circuit breakers
   • Monitoring

Performance (1)

9. Batch Processing (700+ lines)
   • Concurrent processing
   • generate_batch/2 usage
   • Configuration options
   • Error handling
   • Performance optimization
   • Real-world patterns
   • Telemetry monitoring

Index

10. Guides README (250+ lines)
    • Complete guide index
    • Learning paths
    • Use case mappings
    • Best practices summary

Examples (7 total)

Production Examples (7)

All examples are complete, runnable scripts with comprehensive documentation.

1. Resume Parser (550+ lines)

   • Extract structured data from resumes
   • Complex nested schemas
   • Union types for optional fields
   • Format validation
   • Phoenix/Ecto integration patterns

2. E-commerce Product Categorization (500+ lines)

   • Product data extraction
   • Category taxonomy
   • Feature extraction
   • Price tier estimation
   • Tag generation

3. Customer Support Triage (650+ lines)

   • Ticket classification
   • Priority detection
   • Urgency indicators
   • Sentiment analysis
   • Department routing

4. Document Metadata Extraction (600+ lines)

   • Academic paper metadata
   • Nested author objects
   • Format validation (DOI, dates, URLs)
   • Keyword extraction

5. Classification Example (300+ lines)

   • Basic enum classification
   • Sentiment analysis
   • Confidence scoring
   • Batch processing example

6. Ecto Schema Adapter Example (250+ lines)

   • Converting Ecto schemas
   • Automatic validation
   • Integration patterns

7. Customer Support Triage (Original) (600+ lines)

   • Original triage implementation
   • Multiple examples
   • Validation failure examples

Interactive Tutorials

Livebook Notebooks (14 complete)

Beginner Level (1)

1. Getting Started (400+ lines)
   • Introduction to ExOutlines
   • Basic schema creation
   • Validation fundamentals
   • Error handling
   • Backend configuration

Intermediate Level (4)

3. Named Entity Extraction (700+ lines)

   • Pizza order processing example
   • Nested object schemas
   • Handling missing information
   • Batch processing
   • Phoenix integration patterns

4. Dating Profile Generation (700+ lines)

   • Creative content generation
   • EEx template systems
   • Different writing styles
   • Profile quality metrics
   • A/B testing variants
   • LiveView integration

5. Question Answering with Citations (750+ lines)

   • Trustworthy Q&A systems
   • Citation extraction and formatting
   • Source verification
   • Handling insufficient evidence
   • Citation quality validation
   • Knowledge base integration

6. Sampling and Self-Consistency (600+ lines)

   • Multi-sample generation
   • Answer distribution analysis
   • Self-consistency technique
   • Entropy calculation
   • Model comparison
   • Interactive exploration
   • VegaLite visualizations

Advanced Level (5)

7. Models Playing Chess (700+ lines)

   • Constrained chess move generation
   • Dynamic schema generation
   • Legal move validation
   • Game state tracking
   • Interactive chess game

8. SimToM: Simulation Theory of Mind (750+ lines)

   • Two-stage perspective-taking
   • Belief tracking
   • Character knowledge filtering
   • Mermaid diagrams
   • Cognitive reasoning

9. Chain of Thought Reasoning (800+ lines)

   • Step-by-step reasoning
   • Reasoning chain validation
   • Conclusion extraction
   • Mathematical problem solving
   • Complex reasoning patterns

10. ReAct Agent (900+ lines)

    • Thought-Action-Observation loops
    • Tool integration (Wikipedia, calculator)
    • Multi-step problem solving
    • Agent architecture patterns
    • Error handling and recovery

11. Structured Generation Workflow (800+ lines)

    • Multi-stage pipelines
    • Iterative refinement
    • Quality validation
    • Pattern testing
    • Workflow orchestration

Vision and Document Processing (4)

12. PDF Reading with Vision Models (950+ lines)

    • PDF to image conversion
    • Invoice extraction
    • Research paper metadata
    • Multi-page processing
    • Form digitization
    • Document classification

13. Earnings Reports Extraction (1050+ lines)

    • Financial data extraction
    • Income statement parsing
    • Balance sheet validation
    • Multi-period comparison
    • CSV export for analysis
    • Accounting equation validation

14. Receipt Digitization (1000+ lines)

    • Receipt image processing
    • Line item extraction
    • Expense categorization
    • Batch processing
    • Quality assessment
    • Restaurant receipt analysis

15. Extract Event Details (700+ lines)

    • Natural language to structured events
    • Relative date parsing
    • ISO 8601 conversion
    • iCalendar generation
    • Time zone handling

Livebook Index

16. Livebook README (400+ lines)
    • Setup instructions
    • API key configuration
    • Running notebooks
    • Troubleshooting
    • Cost considerations
    • Learning paths
    • Best practices

Main Documentation

1. README.md (630+ lines)

   • Project overview
   • Quick start
   • Installation
   • Usage examples
   • Feature list
   • Backend documentation
   • Contributing guidelines

2. DOCUMENTATION_INDEX.md (This file)

   • Complete documentation inventory
   • Content organization
   • Quick navigation

Statistics

Documentation Coverage

• Guides: 11 guides (6,250+ lines total)
• Examples: 7 production examples (3,550+ lines total)
• Livebooks: 14 interactive notebooks (11,000+ lines)
• Total Lines: ~21,000+ lines of documentation and code examples

Topics Covered

Core Concepts:

• Structured generation
• Schema design
• Validation
• Retry-repair loop
• Backend architecture
• Telemetry

Schema Features:

• String validation (length, regex, format)
• Numeric constraints (min, max)
• Arrays with constraints
• Nested objects
• Union types
• Enum types
• Required/optional fields

Integration:

• Phoenix controllers
• LiveView patterns
• Oban background jobs
• Ecto schemas
• Caching strategies

Testing:

• Mock backend
• Integration tests
• Property-based testing
• CI/CD patterns

Production:

• Error handling
• Retry strategies
• Graceful degradation
• Circuit breakers
• Monitoring
• Batch processing
• Performance optimization

Use Cases:

• Resume parsing
• Product categorization
• Customer support triage
• Document metadata extraction
• Content classification
• Sentiment analysis
• PDF document extraction
• Financial data extraction
• Receipt digitization
• Event information extraction
• Chess game implementation
• Theory of Mind reasoning
• Chain of Thought prompting
• ReAct agent systems
• Multi-stage workflows

Vision and Document Processing:

• PDF to image conversion
• Invoice extraction
• Research paper parsing
• Financial statement analysis
• Receipt digitization
• Form processing
• Multi-page document handling

Content Quality Standards

All documentation follows these standards:

• No emoji (per user requirement)
• Clear, professional style
• Straightforward language
• Complete, runnable code examples
• Real-world use cases
• Error handling examples
• Integration patterns
• Best practices
• Production-ready patterns

Navigation Guide

By User Type

New Users:

1. Getting Started
2. Schema Patterns
3. Classification Example
4. Sampling Livebook

Phoenix Developers:

1. Getting Started
2. Phoenix Integration
3. Ecto Schema Adapter
4. Error Handling

Production Engineers:

1. Core Concepts
2. Architecture
3. Error Handling
4. Batch Processing
5. Testing Strategies

Advanced Users:

1. Architecture
2. Batch Processing
3. Schema Patterns
4. All production examples

By Use Case

Building a Classification System:

• Classification Example
• Schema Patterns (Enum section)
• Testing Strategies

Processing User-Generated Content:

• Resume Parser
• Document Metadata
• Batch Processing
• Error Handling

Real-time Web Features:

• Phoenix Integration (LiveView)
• Error Handling
• Testing Strategies

Background Processing:

• Phoenix Integration (Oban)
• Batch Processing
• Error Handling

E-commerce Applications:

• Product Categorization
• Schema Patterns
• Batch Processing

Contributing to Documentation

Missing Topics

Potential future documentation:

1. Performance Guide: Detailed performance optimization
2. Streaming Guide: Streaming responses (when implemented)
3. Custom Backends: Building custom backend implementations
4. Prompt Engineering: Effective prompt design for structured output
5. Cost Optimization: Minimizing LLM API costs
6. Security Guide: API key management, input sanitization
7. Deployment Guide: Production deployment patterns
8. Monitoring Guide: Comprehensive telemetry and alerting

How to Contribute

1. Identify documentation gap
2. Check existing issues/PRs
3. Follow style guide:
   • No emoji
   • Clear, professional style
   • Complete code examples
   • Real-world patterns
4. Submit PR with updates to this index

License

All documentation is distributed under the same license as ExOutlines (MIT).

Maintained by: ExOutlines Contributors

Repository: GitHub

Questions: GitHub Issues