📘 Guide — Paths & Sections
Choose a path and read in order or jump to any chapter.
Beginner — AI Fundamentals
From AI basics to tokens, embeddings, context, prompts, and LLMs
- 🤖→
1/24
What is Artificial Intelligence?
Fundamentals of AI and how it differs from traditional programming
- 🧠→
2/24
Understanding Neural Networks
How artificial neurons work together to process information
- 📈→
3/24
How Neural Networks Learn
Forward pass, loss, and backpropagation
- 🎯→
4/24
Why You Must Understand Tokens & Embeddings
The case for learning the basics
- 📖→
5/24
The Journey From Text to AI
From raw text to vectors step by step
- 🔬→
6/24
What Are Tokens?
How AI breaks down text into processable units
- ✂️→
7/24
Types of Tokenization
Character, word, and subword tokenization
- 🧹→
8/24
Normalization (Cleaning the Mess)
Making text consistent for the model
- ⚙️→
9/24
Popular Tokenization Algorithms
BPE, WordPiece, SentencePiece
- 💰→
10/24
Why Tokens Matter for Cost
You pay per token
- 📐→
11/24
What Is an Embedding?
Meaningful vectors from token IDs
- ➕→
12/24
Vector Math = Meaning Math
Distance and arithmetic in vector space
- 🔍→
13/24
Vector Search (How RAG Works)
Retrieval Augmented Generation pipeline
- 📐→
14/24
Cosine Similarity
Industry standard for similarity
- 🚀→
15/24
ANN & HNSW
How vector search scales
- 🪟→
16/24
Understanding Context Windows
Memory limits of LLMs and why they matter
- ✍️→
17/24
Your First Prompts
Write effective prompts that get better results
- 🌟→
18/24
Meet the Major LLMs
Overview of GPT, Claude, Gemini, and other popular models
- 🎚️→
19/24
Understanding Model Parameters
What temperature, top-p, and other settings do
- 🔢→
20/24
Open Source & Parameters
Ollama, and what billions of parameters mean
- 🏋️→
21/24
Training a Model
How models learn from data
- ⚠️→
22/24
LLM Limitations & Safety
What LLMs can and cannot do, and common pitfalls
- 🎯→
23/24
Final Mental Model
The full AI stack from text to generation
- 📚→
24/24
Resources: Training, TensorFlow, Hugging Face, Kaggle
Libraries, models, and datasets for training and experimentation
Using AI & AI Agents
Agent components, memory, prompting, tools, MCP, and agentic patterns
- 🤖→
1/24
What Are AI Agents?
Overview: agents that plan, use tools, and act
- 📋→
2/24
Types of Agents & When to Use Them
Simple reflex to multi-agent; open-ended and multi-step use cases
- 🧩→
3/24
Components of AI Agents
Overview: memory, prompting, tools, resources
- 📝→
4/24
System Prompt
Depth: what the system prompt is and how to write it
- 🎯→
5/24
Prompting: Zero-Shot, Few-Shot & Long Context
When to use no examples, few examples, or long context
- ⛓️→
6/24
Prompt Chaining
Multi-step prompts: output of one step becomes input to the next
- 💾→
7/24
Prompt Caching
Reuse cached prefix tokens to cut cost and latency on long prompts
- 🎯→
8/24
Grounding
Anchor model answers in real data: RAG, search, and citations
- 🔄→
9/24
Think and Act
ReAct and agentic loops in depth
- 🧠→
10/24
Memory in Agents
Short-term and long-term memory
- 🔧→
11/24
Tools & Resources
How agents use tools and read resources
- 🛡️→
12/24
Trustworthy Agents & Human-in-the-Loop
Safety, system message framework, threats, and human approval flows
- 📐→
13/24
Planning: Goals and Task Decomposition
Define goals, break into subtasks, structured output, iterative planning
- 👥→
14/24
Multi-Agent Patterns
When to use multiple agents; group chat, hand-off, collaboration; visibility
- 🪞→
15/24
Metacognition in Agents
Thinking about thinking: self-reflection, planning, corrective RAG, code as tool
- 📦→
16/24
Context Engineering
Managing context vs prompt engineering; types, strategies, and common failures
- 📡→
17/24
Introduction to MCP
Model Context Protocol: what and why
- ❓→
18/24
Why We Need MCP
Problems without a standard; one protocol, many apps and servers
- ⚖️→
19/24
MCP vs API
How MCP differs from traditional APIs: discovery, interoperability, and integration
- ⚙️→
20/24
MCP: Client, Server, Host, Resources & Tools
Client, server, host; exposing resources and tools
- 🚀→
21/24
How to Expose Your MCP Server
Build and expose tools, resources, and prompts
- 🤝→
22/24
Agent-to-Agent (A2A) Protocol
Agents talking to agents across systems; agent cards, executor, artifacts
- 🌐→
23/24
Natural Language Web (NLWeb)
Natural language interfaces for websites; MCP, embeddings, and discovery
- 🛠️→
24/24
Resources, Flows & Frameworks
OpenClaw, LangGraph, Cursor/Claude for coding, n8n; good agent flows