AI Jargon
Plain-English definitions of AI buzzwords
AI jargon is everywhere — in meetings, articles, sales pitches. When someone drops “agentic workflow” or “RAG pipeline” and you don’t know what it means, it’s easy to check out of the conversation. This lesson gives you just enough vocabulary to stay engaged and ask smart questions.
Match the jargon
Tap a term on the left, then tap the definition you think matches it on the right. Don’t worry about getting them all right the first time — wrong guesses are part of learning.
Quick reference
That covers the terms you’ll hear most often. The full glossary below includes a few more — bookmark this page and come back when you hear a term you can’t quite place.
| Term | What it means | Where you learned it |
|---|---|---|
| LLM | Large Language Model — the prediction engine behind major AI chatbots | Lesson 1 |
| Token | A word or piece of a word — the basic unit AI reads and generates | Lesson 1 |
| Context Window | The amount of text AI can hold in a single conversation | Lesson 8 |
| Prompt Engineering | Crafting clear, structured instructions to get better AI output | Lessons 4–6 |
| Hallucination | When AI generates confident-sounding but false information | Lesson 7 |
| Agent | AI that takes multi-step actions, not just answering questions | Lesson 10 |
| RAG | Retrieval-Augmented Generation — grounding AI in real documents | Lesson 7 |
| Fine-Tuning | Training a model on specialized data for better domain performance | — |
| Multi-Modal | AI that handles text, images, audio, and more | Lesson 2 |
| Temperature | Controls randomness — low = predictable, high = creative | Lesson 3 |
| Agentic Workflow | A process where AI makes decisions and takes actions with some autonomy | Lesson 10 |
| Connector | A bridge letting AI access your other apps — email, docs, calendar, CRM | Lesson 10 |
| API | Application Programming Interface — how apps talk to each other behind the scenes | Lesson 10 |
| MCP | Model Context Protocol — an open standard for connecting AI to external tools | Lesson 10 |
Takeaway: You don’t need to memorize all of these. The goal is recognition — when you hear these terms in a meeting or article, you now have enough context to follow along and ask good questions.
Not all AI terminology is created equal. Some terms describe real technical concepts that practitioners use every day — tokens, context window, hallucination, RAG, MCP. These are worth knowing because they help you understand what’s actually happening.
Others are marketing language designed to sound impressive. Things like “cognitive AI,” “autonomous intelligence,” or “neural empathy engine” might show up in a product pitch, but they don’t mean much in practice. They’re vibes, not vocabulary.
A good rule of thumb: if you can’t find a clear, concrete definition of a term — one that tells you what’s technically happening — it’s probably marketing. The more specific and boring a term sounds, the more likely it describes something real.
The terms in this lesson are the real ones — the vocabulary that actual AI practitioners use when they’re building, debugging, and evaluating these systems. Master these and you’ll be able to follow any serious AI conversation.
Need help bringing this to your team?
We work alongside your team to build AI-native workflows — from one-week sprints to full engineering acceleration. No handoffs, no slide decks.
Talk to us