Build Applications with Local AI Models on a Mac shows you exactly how to build and run a ChatGPT-style assistant entirely on your own Mac—free forever, with zero data leaving your machine. In this practical book, AI authority Keiji Kamigusa guides you step by step using readily available open source models, Ollama, Python, and Streamlit.
Keiji Kamigusa
If you’ve been curious about running AI locally instead of depending on cloud APIs, this book walks you through building a ChatGPT-style assistant that runs on your own Mac. The focus is practical: install the tools, download open models, write the app, test it offline, and keep improving it.
The book is written for readers who are comfortable using a computer but may be new to programming. It starts with the Terminal, Homebrew, and Ollama, then moves into Python, Streamlit, voice input, chat history, model selection, RAG, and agents.
A few things you’ll build and try along the way:
-
Run Llama, Gemma, Qwen, and Mistral locally with Ollama
-
Create a ChatGPT-like web app with Streamlit
-
Add streaming responses and conversation memory
-
Switch between models from the app UI
-
Use system prompts and temperature settings to shape model behavior
-
Test privacy by turning off the network and watching the app keep working
-
Add document-based question answering with RAG
-
Explore LangChain agents that can use tools
Keiji Kamigusa brings a lot of hands-on experience to the topic. He has worked in AI and machine learning for over 16 years, deployed local language models for more than 50 companies, taught over 42,000 students, and written more than 50 books on AI.
This should be especially useful if you want to experiment with AI apps without sending prompts or documents to a hosted service. It’s also a good fit for Mac users who want a clear path from “I’ve never used the command line much” to “I have a working local AI app.”
Don’t forget you can get 45% off with your Devtalk discount! Just use the coupon code “devtalk.com” at checkout 