Behind every impressive AI application — the assistant that answers from your documents, the tool that takes real actions, the agent that remembers you — sits a small set of building blocks. This series teaches them from nothing and in order, each topic building on the last: embeddings turn meaning into coordinates, retrieval feeds a model the right passages, chunking makes retrieval work, tools give the model hands, MCP standardises the connection, context engineering composes what the model sees, and memory gives it continuity. Every term is spelled out the first time it appears.
-
Article 1
Embeddings & Vector Search, Explained From Zero
How a model turns text into coordinates on a map of meaning, so a computer can find things by what they mean instead of the exact words they contain — the quiet foundation under everything else.
-
Article 2
RAG (Retrieval-Augmented Generation), Explained From Zero
How giving a model the right passages to read at answer-time lets it answer from your private, current documents — accurately and with citations — instead of guessing from memory. Open-book beats closed-book.
-
Article 3
Chunking & Indexing, Explained From Zero
The unglamorous prep that makes retrieval work — why you split documents into right-sized pieces before embedding them, and why chunk size quietly decides whether RAG succeeds or fails.
-
Article 4
Tool & Function Calling, Explained From Zero
How a model that can only produce text learns to look up live data, run calculations, and take actions — by asking your code to call functions on its behalf. The model proposes; your code disposes.
-
Article 5
MCP (Model Context Protocol), Explained From Zero
The standard that lets any AI assistant plug into any tool or data source — the “USB-C for AI” that turns one-off integrations into reusable, shareable connections you build once.
-
Article 6
Context Engineering, Explained From Zero
The discipline of deciding what goes into the model’s limited window — and what to leave out — so it has exactly what it needs to answer well. The right information beats the most information.
-
Article 7
Agent Memory, Explained From Zero
How an assistant remembers things across conversations — your name, preferences, past decisions — even though the model itself forgets everything the moment a chat ends. The layer that makes an agent feel like one.
More in the AI Security collection
-
Companion series
AI Security
Sixteen zero-assumed-knowledge explainers of AI agent security — what an agent is, prompt injection, excessive agency, guardrails, the OWASP Top 10, and MITRE ATLAS.
-
Companion series
Prompt Engineering
The craft of talking to AI models — clear prompts, the message roles that steer a model, teaching by example, chain-of-thought, structured output, and temperature.
-
Companion series
AI Coding Assistants
How GitHub Copilot works and the files that teach it your project — custom instructions, prompt files, instructions files, skills, and custom agents.
-
Companion series
AI Quality & Delivery
Making AI features good and shippable — hallucinations, grounding, evaluations, the prompting-vs-RAG-vs-fine-tuning decision, model selection, and cost.