AI Knowledge

How AI actually works

Not high-level overviews. Practical explanations of the components that make real AI systems work: Agents, RAG, LLMs, embeddings, and architecture — each with working examples.

2 published
6 coming soon

2 of 8 topics published

Agent
RAG
LLM
Architecture
01
AgentBeginnerInteractive demo

What is an AI Agent?

How Agents differ from general AI, what components they need, and how they process real-world tasks step by step.

02
ArchitectureBeginner

Agent Architecture: How It Actually Works

The full picture: how UI, Backend, RAG, Tools, and LLM connect — and why the loop between Backend and LLM is what makes an Agent an Agent.

03
RAGBeginner

RAG: Retrieval-Augmented Generation

How to give AI access to your own data by combining vector search with language models. The foundation of enterprise AI.

Coming
04
LLMBeginner

LLM Prompting Fundamentals

How to communicate effectively with language models: system prompts, few-shot examples, and chain-of-thought reasoning.

Coming
05
RAGIntermediate

Embeddings & Vector Search

The mathematical foundation behind semantic search and how computers represent meaning as numbers in high-dimensional space.

Coming
06
AgentIntermediate

Tool Use & Function Calling

How Agents interact with external systems by calling APIs, querying databases, and executing custom functions.

Coming
07
AgentIntermediate

Agent Memory Systems

Short-term context windows, long-term vector storage, and how Agents learn from past interactions to improve over time.

Coming
08
ArchitectureAdvanced

Multi-Agent Architecture

How multiple specialized Agents collaborate, delegate tasks, and coordinate to solve problems no single Agent could handle alone.

Coming

New topics are added as I study and build. Each entry is written from the perspective of someone who has actually implemented it, not just read about it.