Course Overview
Understand the language of AI—and see how the pieces fit together.
LLMs, AI agents, tokens, RAG, MCP, context engineering... These terms appear everywhere. But what do they actually mean, why do they matter, and how do they connect?
AI Key Concepts & Terms - The Missing Guide gives you the foundation to make sense of modern artificial intelligence. Move beyond recognizing buzzwords and build a clear understanding of the models, tools and systems behind today's AI applications.
Join Maximilian Schwarzmüller and build your understanding through straightforward explanations, diagrams and concrete examples. Whether you are completely new to AI or already use tools like ChatGPT and Claude, you will learn how the key concepts connect.
What you'll learn
Make sense of AI and language models. Understand how AI, machine learning and neural networks relate. Discover what sets LLMs, smaller models and multimodal models apart.
Understand what shapes an AI response. Explore tokens, context windows, temperature, prompt engineering and context engineering. Learn why the information you provide matters.
See how models are trained and improved. Get clear explanations of training, inference, fine-tuning, LoRA, synthetic data, RLHF and distillation. Understand the ideas behind caching, quantization and model routing.
Find out how AI agents get work done. Follow the agent loop and understand the roles of tools and harnesses. Explore MCP, agent skills, computer use and human oversight.
Understand knowledge, memory and reliability. Connect grounding, RAG, embeddings and vector databases. Discover how memory, compaction and context rot affect longer interactions.
Evaluate AI claims and recognize risks. Look beyond benchmark headlines, understand evals, and learn about hallucinations, prompt injection and data exfiltration. Explore open model weights and the possibilities of running models yourself.
A clear starting point. A useful reference.
You do not need programming skills, advanced mathematics or previous AI knowledge. There is no software to install and no paid AI subscription required. The focus is on understanding the concepts through explanations and visual examples.
Most lessons take only a few minutes. Follow the full course to build a connected foundation, or return to an individual lesson when you encounter an unfamiliar term.
Whether you want to follow AI discussions, understand the tools you already use, communicate with technical teams or prepare for more specialized AI learning, this course helps you ask better questions and make sense of what you are hearing.
Join the course and turn AI buzzwords into concepts you can explain with confidence.
FAQ
FAQ area empty
Course curriculum
Chương 1 Welcome & AI Foundations 13 bài · 19 phút
- Welcome to the Course Học thử 1:03
- Download the Complete Course Slide Deck Học thử
- Artificial Intelligence (AI) Học thử 2:24
-
Machine Learning (ML) 2:21
-
Neural Networks & Deep Learning 2:24
-
Large Language Models (LLMs) 3:10
-
Small Language Models (SLMs) 2:24
- Vision Language Models (VLMs) Học thử 1:27
-
Multimodality 1:31
- Frontier Models Học thử 0:49
-
Artificial General Intelligence (AGI) 1:33
-
AI Foundations: Knowledge Check
- Understanding AI & LLMs Học thử
Chương 2 Language Models, Tokens & Context 9 bài · 12 phút
-
Parameters & Weights 1:35
-
Tokens 2:17
- Tokenizers & Tokenization Học thử 0:40
-
The Context Window 2:48
-
Context 1:43
-
Temperature & Sampling 2:15
-
Mixture of Experts (MoE) 1:10
-
Language Models, Tokens & Context: Knowledge Check
-
Diving deeper into LLM core concepts
Chương 3 Training Models & Improving Efficiency 14 bài · 22 phút
-
Training vs. Inference 1:50
-
Pre-Training vs. Post-Training 2:06
-
Fine-Tuning 1:41
-
LoRA & Parameter-Efficient Fine-Tuning 1:38
-
Synthetic Data 2:17
-
Reinforcement Learning from Human Feedback (RLHF) 2:27
-
Distillation 1:51
-
Data Poisoning 1:51
-
Reasoning Models 0:49
-
Caching 1:12
-
Quantization 1:59
-
Model Routing 2:35
-
Training & Efficiency: Knowledge Check
-
Diving deeper into more advanced AI / LLM terms
Chương 4 Prompts, APIs & AI Agents 12 bài · 17 phút
-
Prompts 1:09
-
Prompt Engineering 1:20
-
Context Engineering 1:06
-
Structured Outputs 1:32
-
APIs for AI Models 1:58
-
Models vs. AI Agents 2:23
-
AI Agents 2:18
-
Tools 1:45
-
The Agent Loop 2:32
-
The Harness 1:26
-
Prompts, APIs & Agents: Knowledge Check
-
Discuss tools built on top of LLMs and how they work behind the scenes
Chương 5 Knowledge, Memory & Reliable Answers 9 bài · 17 phút
-
Hallucinations 1:28
-
Grounding 2:10
-
Retrieval-Augmented Generation (RAG) 1:40
-
Vector Databases & Embeddings 2:54
-
Memory 3:26
-
Compaction 2:45
-
Context Rot 2:53
-
Knowledge, Memory & Reliability: Knowledge Check
-
Discuss advanced LLM-related concepts
Chương 6 Agent Capabilities, Oversight & AI Coding 10 bài · 17 phút
-
Model Context Protocol (MCP) 2:43
-
Agent Skills 3:45
-
Computer Use 2:05
-
Sandboxes 2:07
-
Agentic Search 1:59
-
Human in the Loop 1:29
-
Agentic Engineering 1:12
-
Vibe Coding 1:22
-
Agent Capabilities & Oversight: Knowledge Check
-
Discuss how to steer, control and safely use AI and tools built on top of LLMs
Chương 7 Evaluating AI, Risks & Model Choices 7 bài · 10 phút
-
Benchmarks 1:42
-
Evals 1:56
-
Prompt Injection 2:33
-
Data Exfiltration 2:21
-
Open-Weights & Open-Source Models 1:28
-
Evaluation, Risks & Model Choices: Knowledge Check
-
Discuss AI evaluation and risks
Requirment
-
Start from scratch: no prior AI knowledge, programming experience or advanced mathematics required.
-
All you need is curiosity and a device to watch the lessons. No paid AI subscription, API key or software installation required.
Outcomes
-
Explain the AI terminology you encounter every day, from machine learning and LLMs to tokens and context windows.
-
Describe how AI agents work, including tools, agent loops and the software that connects everything.
-
Make sense of model training and fine-tuning, including LoRA, synthetic data, RLHF and distillation.
-
Explain how prompts, context engineering and memory shape AI responses, and why longer conversations can become less reliable.
-
Connect grounding, RAG, embeddings and vector databases to understand how AI works with relevant information.
-
Distinguish MCP, agent skills and computer use, and explain what each adds to AI agents.
-
Recognize hallucinations, prompt injection and data exposure risks, and explain where sandboxes and human oversight help.
-
Look beyond headline benchmark scores: understand evals and the trade-offs behind model size, speed, cost and local deployment.
0 Student
8 Courses
0 review