AI basics for
non-technical people.
You do not need computer science, advanced mathematics or coding experience. Start by understanding the language—models, prompts, context, tools, agents and harnesses—then decide how technical you want to become.
Seven words remove most of the confusion.
Learn these before choosing tools or courses. They describe almost every modern AI workflow.
The pattern engine that produces a response. Different models trade speed, cost, knowledge and reasoning ability.
Your instruction. Good prompts explain the goal, context, constraints and what a useful answer should look like.
A small piece of text a model reads or writes. Token limits affect how much material fits into one conversation.
Everything the model can currently see: your message, previous messages, documents, tool results and instructions.
An ability outside the model—searching the web, reading a file, calculating, running code or querying a database.
A model working toward a goal through multiple steps. It can decide what to inspect, use tools and revise its work.
The working environment around a model. It supplies instructions, tools, file access, memory, safety rules and the loop that lets an agent act.
Read the complete harness guide →Learn in the right order.
Each stage answers one essential question and gives you something concrete to do next.
Use AI without being misled by it.
Begin with a real session. Learn how to state a goal, provide useful context, iterate, recognize uncertainty and check important claims.
Turn plausible answers into grounded research.
Language models predict convincing text. Serious work requires sources, comparisons and independent verification.
Understand tools, agents and harnesses.
A chatbot answers once. An agent can inspect, act, check results and continue. The harness is the system that makes this controlled loop possible.
Move from concepts to a real workflow.
Install a terminal environment, run models, research with sources and see how an agent works with tools and local files.
npm i -g expert-aiOpen installation guide ↗
35 MINExpert AI Terminal product deep dive→
COMPAREModels, access and pricing→
COLLECTIONAI CLI resources→
PACKAGEExpert AI on Pi↗
PACKAGEExpert AI on npm↗
Want to build AI—not only use it?
Follow this branch when you are ready for programming, computer science, machine learning and production systems.
Programming literacy
Learn variables, functions, data structures, debugging, Git and the command line.
Computer science
Study algorithms, data structures, systems, databases, networking and software design.
Machine learning
Understand datasets, training, evaluation, overfitting, probability and optimization.
Deep learning & LLMs
Build neural networks, understand transformers, retrieval, fine-tuning and evaluation.
Agent engineering
Design reliable tool use, state, memory, permissions, observability and evaluation.
Established courses. Zero paid-course clutter.
External course links are limited to respected universities or mature open-source education projects. Choose one route; do not attempt all of them at once.
A globally accessible introduction for people without coding or advanced mathematics.
History, ethics and the social questions surrounding artificial intelligence.
A rigorous introduction to programming designed for learners with or without prior experience.
Computer science and programming fundamentals for majors and non-majors alike.
A complete self-taught computer-science education assembled from open university materials.
Algorithms behind search, classification, optimization, machine learning and language.
A respected, application-first route for people who already have some coding experience.
Read the ideas behind the tools.
This is the technical shelf—not required for beginners. Start with the summaries, then read papers when a concept becomes relevant to your work.
Foundations are a launchpad.
Explore applications in research, architecture and knowledge systems—or return to the main learning map.