Course Overview
This course teaches how to run open AI models entirely on your own machine without relying on API keys, subscriptions, or cloud services. You will learn to build real workflows that process private files locally, understand model cards, choose model parameters deliberately, and evaluate local AI against cloud solutions. The course includes practical labs and assignments to set up a local AI server and develop business ideas using local AI.
Key Takeaways
- Run open AI models locally with no data leaving your computer.
- Read and interpret Hugging Face model cards to assess hardware compatibility.
- Select model parameters such as size, quantization, and file format intentionally.
- Convert your laptop into a local AI server accessible via OpenAI-compatible endpoints.
- Build repeatable workflows that transform private files into usable outputs.
- Perform real evaluations comparing local and cloud AI, and design hybrid workflows.
- Apply a five-part filter to identify workflows benefiting from local AI deployment.
- Explore three local AI business ideas and develop your own vertical strategy.
Prerequisites
- A Mac, Windows, or Linux computer with at least 8 GB of RAM (16 GB recommended).
- An internet connection to download models initially; no internet needed afterward.
- No coding background or paid subscriptions required.
Target Learners
- Founders and operators who believe local AI is inaccessible to them.
- Knowledge workers with sensitive files that cannot be uploaded to cloud services.
- Consultants and agency owners targeting industries using outdated software.
- Anyone wanting AI to process their own files locally without data leaving their premises.
- Not intended for machine learning engineers seeking training or fine-tuning skills.
What You'll Build, And What Local AI Means 4:43
Local Or Cloud — Where Should Intelligence Live? 6:31
The Better Question: Is It Good Enough For The Job? 5:11
The Four Pieces 6:13
Section 1 Quiz — Why Local AI
Parameters And Model Size — And What The 'E' Means 5:00
Tokens And The Context Window 5:22
Quantization Without The Maths 5:13
File Formats: GGUF, safetensors, MLX, LiteRT-LM 4:56
Lab: Reading A Model Card Slowly 4:49
Section 2 Quiz — The Vocabulary Decoder
Lab: Hugging Face, The Model Warehouse 4:54
The Gemma Family And How To Pick A Size 6:01
The Specialized Gemmas 5:20
Llama, Qwen, DeepSeek, GLM, Mistral, Phi 5:28
Licences, Procurement, And Open Weights vs Hosted 5:23
Section 3 Quiz — The Open Model Landscape
The Hardware Cheat Sheet 5:40
Lab: LM Studio — Install To First Real Answer 4:50
Lab: LM Studio's Local Server — Your Laptop As An AI Server 4:48
Lab: Ollama — Pull, Run, And The API On 11434 5:08
Under The Hood: llama.cpp And MLX 5:12
On-Device: AI Edge, LiteRT-LM, Edge Gallery 5:11
Section 4 Quiz — Three Ways to Run a Model
Assignment A — Your First Local Answer, and a Hardware Plan
The Folder Pattern: One Folder, One Model, One Output 4:58
Lab: Building The Customer-Notes Reviewer 4:39
Lab: Prompt, Examples, Checklist 4:44
Artifacts Beat Chat Answers 4:43
Lab: Run It Ten Times — Finding Where It Breaks 4:39
Section 5 Quiz — Your First Local Workflow
Assignment B — The Customer-Notes Reviewer
What An Eval Actually Is 4:50
Lab: A Real Eval On The Same Ten Files 4:38
The Hybrid Pattern: Local, Cloud, Human 5:25
Privacy, Residency, And What Never Leaves 5:13
Workflows Before Fine-Tuning 4:56
Section 6 Quiz — Evals, Hybrid, and Knowing When to Stop
Assignment C — The Local vs Cloud Eval Scorecard
The Opportunity Filter 5:30
Idea 1 — A Local QA Reviewer For Home Health 5:11
Idea 2 — An Offline Field-Report Copilot 5:09
Idea 3 — A Pre-Send Reviewer For Professional Services 5:18
Service First, Product Second 4:31
The 24-Month Window — Picking Your Vertical 5:17
Section 7 Quiz — Where the Businesses Are
Lab: Setting Up Your Local AI Lab Folder 4:46
Lab: Capstone Walkthrough — Ten Files, One Artifact 4:35
Lab: Local Embeddings And Search Over Your Files 4:36
Where To Go Next 4:48
Section 8 Quiz — Capstone
Assignment D — Capstone: Your Local AI Lab
Final Practice Exam — Local AI (40 questions)
Access Files No pi required.
Free Course files 1 file package
- Full Pack
Discussions are closed
Comments are currently disabled for this course.
