Project: AI Education & Empowerment
The Instructor's Perspective
We are living through a fundamental shift in how humans interact with knowledge. My goal with this project is to demystify AI, moving it from a “black box” to a “force multiplier” that every student and professional can use safely and effectively. We focus on Local First to ensure privacy, autonomy, and low-cost access.
Objective
To create simple, repeatable educational tools and documentation that help people understand the power of LLMs while maintaining strict “signal discipline” regarding ethics and security.
The Strategy: Local-First Hybrid
The AI PACE Plan
P (Primary): Local LLMs (Ollama, LocalAI) running on home-lab hardware or personal workstations. A (Alternate): Privacy-focused APIs (Claude, OpenAI) for high-complexity tasks. C (Contingency): Pre-computed local knowledge bases and vector stores. E (Emergency): Human expertise and physical reference materials.
Key Deliverables
- Ollama Setup Guide: A simple SOP for running LLMs on consumer hardware (SYCL-optimized).
- Prompt Engineering for Students: A guide on how to ask “stupid questions” to get smart answers from AI.
- AI Safety SOP: Best practices for protecting sensitive data while using LLMs.
- Hybrid Solution Architectures: Documentation on connecting local LLMs with external tools (like this Obsidian Vault!).
Current Collaborators
- Garth Johnson (The NetYeti)
- Gemini CLI (AI Assistant & Force Multiplier)
- Cascade STEAM Service Corp
Related: AI Command Center, The Stack, Policies & SOPs