# Ship Local AI with Ollama
Build 5 Real Projects on Your Own Hardware in 30 Days

- Price: $12.99 (Kindle ebook)
- Level: Beginner
- Pages: 193
- Best for: Privacy-first local AI builders
- You finish with: 5 real Ollama projects on your own GPU in 30 days
- Series: Local AI Playbooks, book 1 of 2
- Buy on Amazon: https://amzn.to/3NWMkrJ
- Web page: https://youcanbuildthings.com/books/local-ai-ollama

## What you get

Stop paying $200/month for API calls. Run GPT-class models on your own machine. No API key, no monthly bill, no data leaving the room. 5 real projects from VS Code AI assistant to RAG knowledge base to freelance service.

Run GPT-class models on your own hardware. 5 real projects from coding assistant to RAG knowledge base to freelance offering. 193 pages, zero API bills.

## Chapters

1. $200/Month in API Bills, or Zero: The real cost comparison that makes local AI obvious.
2. The Hardware Truth: What you actually need to run models locally.
3. Pick Your Weapon: Ollama, LM Studio, and the model zoo compared.
4. Zero to Running in 15 Minutes: Install, pull a model, start chatting.
5. Your First AI Pair Programmer: A free coding assistant in VS Code.
6. Build Your Private Knowledge Base: Local RAG with ChromaDB and Ollama.
7. Connect Everything: API server, chatbot, home server setup.
8. Lock It Down: Privacy and security for local deployments.
9. Agents, Tool Calling, Multi-Model Routing: Advanced patterns that make local AI production-ready.
10. The Money: Freelance, consult, and sell local AI services.
11. The Honest Chapter: What local AI still can't do.
12. Your Next Move: Where local AI is headed and how to stay ahead.

## Free guides from this book

- [How to Build a Local RAG System with Ollama and ChromaDB](https://youcanbuildthings.com/articles/build-local-rag-ollama-chromadb-python.md): Build a private document Q&A system using Ollama, ChromaDB, and Python. Runs on 8GB RAM with zero API costs, full data privacy, and no cloud dependency.
- [How to Replace GitHub Copilot with Ollama in VS Code](https://youcanbuildthings.com/articles/free-ai-coding-assistant-vscode-ollama-cline.md): Set up a free AI coding assistant in VS Code using Ollama and Cline. Seven real workflows tested with honest quality scores compared to GitHub Copilot.
- [How to Route Queries to Multiple Ollama Models](https://youcanbuildthings.com/articles/ollama-multi-model-routing-python.md): Build a Python router that sends coding questions to CodeLlama and general queries to Llama 3. Keyword-based and AI-powered routing strategies compared.
- [How to Secure Your Ollama Installation](https://youcanbuildthings.com/articles/ollama-security-hardening-audit-script.md): Run this 30-second audit script to catch the exact misconfiguration that exposed thousands of Ollama instances to the public internet. Copy-paste fixes included.
- [Ollama vs ChatGPT: Honest Benchmarks from 100 Prompts](https://youcanbuildthings.com/articles/ollama-vs-chatgpt-local-ai-benchmarks.md): Benchmark scores comparing local Ollama models against GPT-4o across coding, writing, reasoning, and more. Data from 100 test prompts on a consumer laptop.