← Back to homeJacksonville Urban League

AI Crash Course Cheat Sheet

Jacksonville Urban League · Presented by MyVillage Project Technologies & Coding in Color

Referral Code: JULAI2026

Tools & Platforms

Text & Chat

ChatGPT

Conversational AI by OpenAI

chatgpt.com

Claude

Conversational AI by Anthropic

claude.ai

Google Gemini

Google's AI assistant

gemini.google.com

Perplexity

AI-powered search engine

perplexity.ai

MyVillageOS

Community AI platform

myvillageproject.ai

Audio & Music Generation

ElevenLabs

Realistic text-to-speech

elevenlabs.io

Suno AI

AI music generation

suno.com

App & Design Builders

Lovable

AI website & app builder

lovable.dev

Replit

Online IDE with AI assistance

replit.com

Canva

Design & visual content

canva.com

AI Models & Automation

HuggingFace

Open-source AI models

huggingface.co

n8n

Workflow automation platform

n8n.io

Zapier

App integrations & automation

zapier.com

Key Vocabulary

Model
A piece of code that takes an input and predicts an output. Can be as simple as autocomplete on your phone or as complex as recommending products on Amazon.
Training
Teaching a model by showing it many examples of input and the correct output. Like training a dog — more repetitions with good data means better performance.
Large Language Model (LLM)
A model trained on massive amounts of text that can take in language and generate language back. ChatGPT, Claude, and Gemini are all powered by LLMs.
Small Language Model (SLM)
A smaller, more focused version of an LLM. Designed to run on less powerful hardware or handle specific tasks efficiently.
Token
A chunk of text — roughly 4 characters or about one word. Tokens are how AI measures input and output. Pricing and usage limits are counted in tokens.
Input & Output
Input is what you send to the AI (your question, prompt, or data). Output is what the AI sends back. Everything in AI flows through this pattern.
Context Window
The amount of text a model can "see" at one time. The AI re-reads the whole conversation each turn. When the window fills up, it starts losing context.
Agent
An AI that can take action — not just answer questions. It can send emails, book flights, schedule events, or analyze documents on your behalf.
Connectors / Tools
Integrations that give an agent access to external services — your email, calendar, Canva, the web, etc. Without tools, an agent can only talk.
Skills / Memory
Files and instructions you give an AI agent so it knows who you are, how you work, and what you care about. More context means better performance.
MCP (Model Context Protocol)
A standard that lets AI models connect to external tools and services. Think of it as the plug that lets a model actually do things beyond generating text.
API (Application Programming Interface)
A way for different software systems to talk to each other. When an app uses AI behind the scenes, it's usually calling an API.
Generative AI
AI that creates new content — text, images, audio, code — rather than just classifying or sorting existing content.
Robotics
Physical devices — from robotic arms to edge sensors — that connect to AI models to make decisions and take action in the real world.
Markdown File
A simple text file format used to write structured documents. In the AI world, markdown files are how you create "skills" and reference documents that agents can read.
Data
The raw material used to train models. Could be images, text, numbers, or purchase history. The quality and quantity of your data determines how good your model will be.

Token Economics

AI services charge by tokens. Free plans and subscriptions are really about token allotments. Hit your limit and you pay more or get cut off.

Monthly Subscription

Flat fee (e.g., $20/mo) with a set token allowance. Simple and predictable.

Per-Token Usage

Pay as you go (e.g., $3/M input tokens, $15/M output tokens). Scales with usage.

Fun fact

The average person processes ~40M words/year (speaking, hearing, reading, thinking, writing). At ~$1/M tokens, that's roughly $40 worth of AI processing — compared to $50K–$100K in salary for a knowledge worker.

Key Themes

1

AI as a life skill, not just a tool

Framed through the food/gardening analogy — understand how it works, not just how to use it.

2

Models are not new

Text prediction, recommendations, and classification have been around for years.

3

Tokens = the currency of AI

Understanding input/output economics is foundational to using AI wisely.

4

Agents are the future

Moving from asking questions to delegating tasks.

5

Own your pipeline

The long game is hosting your own models, data, and agents.

6

Security in perspective

Your phone already knows everything. Be smart, but don't freeze.

7

Community-rooted

Everything ties back to how villages and families can use this.