Topics
Long-lived knowledge pages, kept current. Each page cites the source notes it draws on.
Fundamentals
- EmbeddingsAn embedding is a vector (a list of numbers) that represents the meaning of a piece of text, image, or other data. Items with similar meaning end up close together in vector space, so "find…
- Large language modelsA large language model (LLM) is a neural network trained to predict the next token in a sequence of text. Trained on a huge corpus and then tuned to follow instructions, that one skill turns into…
- Reasoning models and extended thinkingReasoning models generate internal reasoning ("thinking") before their final answer. Spending more compute at inference time this way improves results on maths, coding, planning, and multi-step…
- Retrieval-augmented generation (RAG)RAG answers a question by first retrieving relevant information from an external source and then giving it to the model as context. It keeps answers current, grounded, and citable without…
- Tokens and context windowsModels read and write tokens, not characters or words. A tokenizer splits text into sub-word pieces from a fixed vocabulary. Everything about using an LLM is measured in tokens: price, speed, rate…
- Training and fine-tuningLLMs are built in stages. Pre-training gives broad knowledge and language skill; post-training turns the result into a helpful, safe assistant. Fine-tuning adapts a trained model further for a…
- TransformersThe transformer is the neural network architecture behind nearly every modern LLM. It was introduced in 2017 for translation and replaced recurrent networks because attention can process a whole…
Models
- Claude (Anthropic)Claude is Anthropic's family of models, offered through claude.ai, the Claude API, Claude Code, and cloud platforms (Amazon Bedrock, Google Vertex AI, Microsoft Foundry, and Claude Platform on…
- Gemini (Google)Gemini is Google DeepMind's model family, used in the Gemini app, Google products, and through the Gemini API and Vertex AI. Gemini is natively multimodal (text, images, audio, video) and known…
- GPT (OpenAI)OpenAI's GPT models power ChatGPT, Codex, and the OpenAI API. GPT-5 (August 2025) merged the separate GPT and "o-series" reasoning lines into one family. A series of GPT-5.x releases followed…
- Open-weight modelsOpen-weight models publish their trained weights, so anyone can download, run, fine-tune, and self-host them. "Open-weight" is more accurate than "open-source": training data and code are usually…
Agents
- Agent patternsAn agent is an LLM that works in a loop: it decides what to do, uses tools, observes the results, and continues until the task is done. Agents suit open-ended tasks whose steps cannot be…
- Memory for agentsModels have no memory between requests: each call sees only what is in its context. Agents that work for long periods or across sessions need ways to keep what matters and drop what does not.…
- Tool use (function calling)Tool use lets a model call functions you define: search a database, call an API, run code, edit files. The model does not run anything itself. It returns a structured request (tool name plus JSON…
Tooling
- Agent skillsA skill is a folder of instructions, scripts, and reference files that teaches an agent how to do a specific job: fill in a PDF form, follow a company style guide, run a deployment. The agent sees…
- Coding agentsCoding agents read a codebase, plan, edit files, run commands and tests, and iterate until a task is done. They range from IDE assistants to terminal tools to cloud agents that open pull requests…
- Model Context Protocol (MCP)MCP is an open protocol for connecting AI applications to tools and data. A service builds one MCP server; any MCP-capable client (Claude, ChatGPT, IDEs, coding agents, custom agents) can use it.…
Frameworks
- Agent and LLM framework landscapeFrameworks provide building blocks for LLM applications: model clients, tool loops, retrieval, memory, orchestration, tracing. They speed up getting started, but add a layer between you and the…
Practices
- Cost and cachingLLM costs are driven by tokens: input, output (including thinking), and cached input. Most savings come from not reprocessing the same context, choosing the right effort and model for each task,…
- Evaluating LLM applicationsAn eval is a repeatable test of how well a model, prompt, or agent does a task. Without evals, changes to prompts, models, or tools are guesswork. Public benchmarks compare models in general; your…
- Prompting and context engineeringPrompting is writing the instructions and context a model works from. As models have improved, the craft has moved from tricks and magic phrases toward context engineering: choosing what…
- Security for LLM apps and agentsLLM applications add new attack surfaces: the model follows instructions found in any text it reads, and agents can act on those instructions with real tools and credentials. The core rule is to…