Cover Material, Copyright, and License

Preface

1. Requests from the Author
2. Parts of this Book are Specific for macOS and iOS, with Some Support for Linux
3. Code for this Book
4. Author’s Background
5. Cover Art
6. Swift 3rd Party Libraries
7. Acknowledgements

Part 1: Getting Started with Swift for AI Development

Setting Up Swift for Command Line Development

1. Installing Swift Packages
2. Creating Swift Packages
3. Accessing Libraries that You Write in Other Projects

Background Information for Writing Swift Command Line Utilities

1. Using Shell Processes
2. FileIO Examples
3. Mono-Repo Umbrella Package
4. Swift REPL

Web Scraping

1. Running in the Swift REPL

Part 2: Large Language Models

Using the OpenAI LLM APIs

1. Core Architecture
2. Key Features
3. Technical Implementation Details
4. Running Tests

Using Ollama to Run Local LLMs

1. Running the Ollama Service
2. The OllamaService Actor Library
3. Tool Definitions
4. Example Tests
5. Ollama Wrap Up

Running Local LLMs with Apple’s MLX Framework

1. Background: MLX and Apple Silicon
2. Choosing a Model
3. Project Structure
4. Package.swift
5. main.swift — Full Walkthrough
6. Running the Example
7. Swapping Models
8. Key Takeaways
9. Summary

Using the AnyLanguageModel Package with OpenAI, Gemini, and Ollama

1. OpenAI Example
2. Google Gemini Example with Tool Use
3. Ollama Local Example
4. AnyLanguageModel Package Wrap Up

Part 3: Deep Learning, Natural Language Processing, and Retrieval-Augmented Generation

Deep Learning Introduction

1. Simple Multi-layer Perceptron Neural Networks
2. Deep Learning

Natural Language Processing Using Apple’s Natural Language Framework

1. Using Apple’s NaturalLanguage Swift Library
2. NLP Utility Library
3. Running the Example
4. Chapter Wrap Up

Document Question Answering Using Gemini APIs and a Local Embeddings Vector Database

1. Gemini API Client
2. Text Chunking
3. In-Memory Vector Store
4. Running the Example
5. Chapter Wrap Up

Part 4: The Semantic Web, Knowledge Graphs, and Linked Data

Linked Data and the Semantic Web

1. Understanding the Resource Description Framework (RDF)
2. Frequently Used Resource Namespaces
3. Understanding the SPARQL Query Language
4. LLMs and the Semantic Web
5. Semantic Web and Linked Data Wrap Up

Querying Knowledge Graphs with SPARQL and Swift

1. What We Are Building
2. Project Structure
3. The Library: Full Source Listing
4. Running SPARQL Queries
5. The Test Suite
6. Tips for Writing SPARQL Queries
7. Wrap Up

Part 5: Apple Intelligence — On-Device LLMs with FoundationModels

Using Apple Intelligence’s Default System Model To Build a Chat Command Line Tool

Using Apple Intelligence to Build a Coding Assistant

1. Requirements
2. Project Layout
3. Package.swift
4. CodingCLI.swift — Full Listing
5. Walking Through the Code
6. Running the Tool
7. Ideas for Extension
8. Wrap Up

Part 6: Applied AI Projects

Knowledge Base Navigator: Building an AI-Powered Information System

1. Project Overview
2. Project Structure
3. Core Implementation
4. Running the Application
5. Key Takeaways
6. Environment Setup

Anomaly Detection

1. Motivation
2. The Gaussian Model
3. Swift Implementation Overview
4. AnomalyDetection Class
5. Preprocessing the Wisconsin Data
6. Feature Histograms
7. Running the Example
8. Interpreting the Metrics
9. Summary

AutoContext: Prepare Effective Prompts with Context for LLM Queries

1. Project Structure
2. Implementing Vectorization of Text and Semantic Similarity
3. Core AutoContext Implementation
4. The Interactive CLI (main.swift)
5. Example Session
6. Key Takeaways
7. Wrap Up

Book Wrap Up