AI Essentials
The basic building blocks worth understanding before the rest of this site's agent, tool, and retrieval concepts make full sense.
- Context Window — is the maximum amount of text a model can read and respond to at once — everything shares that one budget, including the answer it writes back.
- Fine-Tuning — trains an already-trained model further on examples you supply, changing how it behaves rather than adding to what it knows.
- Hallucination — an AI hallucination isn't a bug — it's the model doing its normal job with nothing solid to base the answer on.
- Reasoning Models — a reasoning model works through a problem in explicit intermediate steps before answering, trading speed and cost for better results on multi-step tasks.
- Temperature — controls how sharply an LLM favors its most likely next token — low values are focused, high values are more random.
- Token — is the small chunk of text a model actually reads and generates — often a whole word, sometimes just part of one, not the same thing as a word.