Glossary

A debt payoff strategy where the borrower attacks the debt with the highest APR first while paying minimums on all other debts. Mathematically optimal for minimizing total interest.

A debt payoff strategy where the borrower attacks the debt with the smallest balance first, regardless of APR. Designed for psychological motivation through quick wins.

A budgeting method where every dollar of monthly income is assigned a specific job (spending category, savings goal, or debt payment) before the month begins. Income minus assignments equals zero.

Interest calculated on both the original principal and accumulated interest. Works against you on debt and for you on savings/investments.

The smallest amount required to pay each month to keep the account in good standing. Paying only the minimum traps borrowers in long-term debt.

A credit card with a 0% intro APR offer for transferred debt. Used to pause interest accrual during accelerated payoff. Typically 3-5% transfer fee.

A budgeting method where each spending category gets a separate "envelope" of money. When the envelope is empty, you stop spending in that category.

US federal law (effective 2022) that protects patients from most surprise out-of-network medical bills for emergency care and in-network facility care.

The Annual Percentage Rate charged on credit card balances. Compounded monthly (or daily on some cards), making it the dominant cost factor in carrying credit card debt.

The practice of crafting effective text prompts to get desired outputs from large language models like ChatGPT, Claude, or Gemini. Includes techniques like role assignment, few-shot examples, and structured output formatting.

A reusable prompt structure with variable placeholders (e.g., {{topic}}, {{audience}}) that can be filled in for different specific use cases. Allows one base prompt to generate hundreds of variations.

A customized version of ChatGPT created with specific instructions, knowledge files, and tools. Built via the GPT Builder by ChatGPT Plus subscribers.

The basic unit of text a language model processes. One English word averages 1.3 tokens. Token counts determine API costs and context window limits.

An AI architecture where the language model retrieves relevant documents from a knowledge base before generating a response. Reduces hallucinations.

The maximum amount of text (measured in tokens) a language model can process in a single request. Larger windows allow for longer documents and conversations.

When a large language model generates plausible-sounding but factually incorrect information presented with high confidence. A known limitation of current LLMs.

An open protocol that lets websites instantly notify search engines about new or updated content. Supported by Bing, Yandex, Naver, and Seznam.

Most US food date labels are quality indicators set by manufacturers, not federal safety dates. Only infant formula has a federally regulated use-by date.

A food storage practice where older items are positioned in front of newer items so they get used first. Reduces waste from forgotten food going bad in the back of the pantry or fridge.

Surface dehydration of frozen food caused by air contact and temperature fluctuations. Visible as white or gray patches with dry texture. Per USDA, freezer-burned food is safe to eat but quality is reduced.