AI Place Hub

Published

Collections

• 4 items

  First week with a coding model

  Read these in order if you only just started using a model on a real codebase. Four entries, about an hour.

• 3 items

  Going local

  How to stop paying for an API: the hardware maths, the install, and the two tools that matter.

• 5 items

  Set up your AI once

  Stop retyping the same rules. Five lessons on moving them into a config file for Claude Code and Cursor, and on picking a model…

• 5 items

  AI on a budget

  Five free lessons, from the free ways to use AI through choosing a model to running one on your own computer. You spend nothing.

• 6 items

  Prompt engineering: zero to fluent

  Six lessons. You start with what a single prompt is made of and end with few-shot examples and a self-critique pass. No…

• 5 items

  AI for study and everyday work

  Five free lessons on turning a messy task into a clear prompt, getting summaries you can trust, and studying without handing the…

• 5 items

  Use AI without getting burned

  Five free lessons about checking answers, keeping private data out of chats and dealing with prompt injection. No sign-up.

• 5 items

  Work with a coding assistant

  Five free lessons: how to brief an assistant, where to keep your rules, and how to review what it changed.

Glossary contributions

• Agent

  A model with tools and a goal that picks its own next steps in a loop and does not answer just once.

• API key

  A secret string that identifies you to an AI service and lets your code call it. Whoever has the key can spend your money.

• Chain of thought CoT

  Prompt the model to write out its intermediate steps before giving the final answer.

• Connector

  A link between an AI app and an outside service, such as a drive, calendar or chat tool, that lets the model read your data from there.

• Context window

  The most text, counted in tokens, that a model can consider at once. Your prompt and its answer share that space.

• Embeddings

  Lists of numbers that stand for text, built so that similar meanings end up close together. Semantic search runs on them.

• Few-shot prompting

  Put two or three worked examples of the task in the prompt, on top of the description.

• Fine-tuning

  More training for an existing model on your own examples. It changes how the model behaves, not what it knows.

• Hallucination

  A fluent, confident answer that is plain wrong: an invented citation, function, statistic or fact.

• Inference

  Running a trained model to get output. It happens every time you send a prompt.

• Large language model LLM

  A model trained on huge amounts of text to predict the next token. That one skill lets it write, summarise, translate and reason over text.

• Model Context Protocol MCP

  An open standard that lets an AI app connect to outside tools and data, such as files, databases or services, through small programs called servers.