Articles tagged "glossary"

What is prompt injection? The 3 attack surfaces to fix

Prompt injection hijacks your LLM via user input, retrieved docs, and tool results. Three attack surfaces, and the defenses that actually work in production.

Sep 16, 2026 · 7 min read glossarysecurityllm

What is an AI agent? Tool loops, stalls, and 3 prod decisions

An AI agent is an LLM in a tool-calling loop. How loops work, why they stall or run away, and the 3 architecture decisions that matter in production.

Sep 7, 2026 · 6 min read glossaryagentstool-calling

LLM hallucinations explained: causes, types, production fixes

Why LLMs hallucinate, the three error types, prompting patterns that reduce them, and the one production eval that catches failures before your users see them.

Aug 28, 2026 · 5 min read glossaryhallucinationsllm

JSON mode explained: when schema enforcement beats prompting

JSON mode vs structured output: how constrained decoding works, when 3% failure rates become a real cost, and which enforcement approach your pipeline needs.

Aug 26, 2026 · 5 min read glossaryjsonstructured-output

System prompt explained: behavior, cost, and pitfalls

A system prompt is the hidden instruction block that shapes every LLM response. Here's how it works, what it costs, and the production pitfalls to avoid.

Aug 3, 2026 · 6 min read glossaryfundamentalsprompt-engineering

Fine-tuning explained: when it beats prompting and when it doesn't

Fine-tuning trains a model on your data to shape behavior and output format. Here's when it beats prompting, what it costs in 2026, and when to skip it.

Jul 20, 2026 · 5 min read glossaryfine-tuningprompt-engineering

Function calling explained: how LLM tools work in production

Function calling lets LLMs trigger your code via JSON schemas. How the loop works, when to use it over plain prompting, and the pitfalls most builders hit.

Jul 10, 2026 · 7 min read glossaryfunction-callingtool-use

Embeddings explained: vectors, RAG, and which model to pick

Embeddings convert text into vectors that capture meaning. Here's how cosine similarity works and how to pick the right model for your RAG pipeline.

Jul 6, 2026 · 6 min read glossaryembeddingsrag

What is MoE? The sparse expert trick behind DeepSeek and Mixtral

Mixture of Experts models run only a fraction of their parameters per token. Here's why DeepSeek and Mixtral are cheap, and when MoE gets expensive.

May 22, 2026 · 7 min read glossaryfundamentalscost

1 token is not 1 word: LLM conversion rates that predict your bill

The exact token-to-word and token-to-character conversion rates for English, code, and non-English LLM input, plus a practical counting recipe.

Apr 27, 2026 · 6 min read glossarytokenscost

What is RAG? The 3 components and when not to use it

RAG has 3 moving parts: ingestion, retrieval, and generation. Here's what each does, when RAG beats fine-tuning, and when to skip it entirely.

Apr 22, 2026 · 6 min read glossaryragfundamentals

Context windows explained: why your 128k model only gives you 100k

The context window is your LLM's working memory per call. What 128k tokens actually fits, why usable size is smaller than advertised, and how to check yours.

Apr 21, 2026 · 7 min read glossaryfundamentalsvibe-coders