🧬AI Engineering
Frameworks for building with LLMs · 36 terms
LLM APIs
11How code calls a language model: the request, what it costs, and the model makers' own APIs.
A web service your code calls to get text back from a big AI model.
Claude's own web service. Your code sends it a chat and gets a reply.
OpenAI's own web service. Your code sends it a request and gets a reply.
A Claude service that runs your agent for you. Its tools run in a sandbox.
How many tokens a call used. Input and output are each counted.
What you pay to use a model. There is a set price for each million tokens.
Send lots of model requests at once. The answers come back later and cost less.
Client SDKs
4Code kits that call a model's API from your own language. One call goes out and one reply comes back.
A kit of code that calls an AI model's API for you, in your own language.
Anthropic's own code kit. Your code calls Claude with one short function.
OpenAI's own code kit. Your code calls its models with one short function.
Google's code kit for Gemini. Your code calls the model with one function.
LLM Gateways
4One door in front of many models. Which model gets each call, and what happens when one fails.
One place that all the AI calls in a firm go through.
A free tool that lets your code call over 100 models in one way.
Sending each request to the model that fits it, cheap or strong.
Sending a call to a backup model when the first one fails.
LLM Frameworks
11Code kits with ready parts for building apps on AI models.
A kit of ready code for making AI apps. You snap the parts together.
A free kit of code for making AI apps from chains of steps.
A free kit of code that hooks AI up to your own files.
A free kit of code for making AI agents as a map of steps.
A TypeScript kit for AI apps. One set of calls works with every model.
A free Python kit. A crew of agents, each with a role, shares one job.
A Python kit for agents. It gives back answers as checked, typed data.
A Python kit. You say what goes in and out, and it writes the prompts.
Agent SDKs
3Code kits that run the whole agent loop for you, calling the model and its tools until the job is done.
A code kit that runs the agent loop for you. It calls the model and the tools.
OpenAI's small kit for agents that use tools and pass on work.
Google's free kit for building AI agents. It works in five coding languages.
LLM App Builders
5Tools where you build an AI app by joining blocks on a screen, and the products that work this way.
A tool for making AI apps by joining blocks on a screen. Little code is needed.
An open-source platform for building AI apps from blocks. Use its cloud or host it.
An open-source drag-and-drop tool for AI agents. Its makers stopped work on it in 2026.
A free tool for making AI agents from blocks. Each block is Python code you can change.
Microsoft's tool for making AI agents with little code. Put them in Teams or on a site.
LLM App Operations
5Running an AI app once it is live. Watching it, saving money and changing prompts with care.
Keeping a record of each step an AI app takes, so you can fix it.
A store that gives back a saved answer when a new question means the same.
Keeping each change to a prompt as a numbered, tested version.