What concepts do these AI flashcards cover?

The deck covers the foundational vocabulary of modern AI: what an LLM is, tokens, prompts, inference, training, and parameters; the architecture (transformer, attention, embeddings, context window); sampling parameters (temperature, top-p, max tokens); prompting techniques (zero-shot, few-shot, chain-of-thought, system prompts); and the building blocks of real systems (RAG, fine-tuning, RLHF, agents, tool use, grounding, alignment). Each card gives a clear, jargon-free definition.

How do I use the flashcards?

Tap a card to flip it between the term and its definition, then use Next and Previous to move through the deck. You can filter by category — Foundations, Architecture, Parameters, Prompting, Techniques, Safety, and more — to drill a specific area. Everything runs in your browser with no account needed.

Are the definitions accurate and beginner-friendly?

Yes. Each definition is written in plain English and reviewed for technical accuracy, with the goal of being correct without assuming prior machine-learning knowledge. Where a concept is commonly confused (for example training versus inference, or temperature versus top-p), the wording is chosen to draw the distinction clearly.

Is this enough to understand AI articles and product docs?

It covers the terms you will meet most often in AI news, tool documentation, and technical discussions, so it removes the biggest barrier — unfamiliar jargon. For deeper understanding of any single concept, pair the flashcards with a focused explainer; the cards are designed to give you the map, not the whole territory.

AI Flashcards: 30 Core Concepts for Beginners

Learn the language of AI

Most of the difficulty in getting started with AI is vocabulary. The ideas are not that hard, but the words — token, embedding, transformer, RAG, inference, RLHF — arrive all at once and make every article feel impenetrable. This interactive flashcard deck fixes that by drilling the core concepts one at a time. Each card shows a term; tap it to reveal a clear, plain-English definition. The deck spans the foundations (what an LLM is, tokens, prompts, training versus inference), the architecture behind the models (transformer, attention, embeddings, context window), the knobs you can turn (temperature, top-p, max tokens), the main prompting techniques (zero-shot, few-shot, chain-of-thought, system prompts), and the building blocks of real applications (RAG, fine-tuning, agents, tool use, grounding, alignment).

How the flashcards work

Tap any card to flip it between the term and its definition, and use Next and Previous to move through the deck. A category selector lets you focus on one area at a time — review just the Architecture cards, or just Prompting, or leave it on All to sweep everything. The cards are written to be accurate and beginner-friendly at the same time, with extra care around the distinctions people most often get wrong: training versus inference, temperature versus top-p, fine-tuning versus RAG. Nothing is sent anywhere — the whole deck runs in your browser with no sign-up, so you can study as often as you like.

Tips for studying effectively

Flashcards work because of active recall: trying to remember the answer before you see it strengthens the memory far more than re-reading does. So before you flip each card, pause and say the definition to yourself, then check. Loop the deck several times and slow down on the cards you keep missing — those are the concepts worth a deeper look elsewhere. Filtering by category helps you isolate a weak area, for example spending a focused session only on the Techniques cards until RAG, fine-tuning, and grounding are crystal clear. Once the terms come to you automatically, test yourself with a quiz to confirm the knowledge holds up under pressure, then start reading real AI documentation — you will find it suddenly makes sense.