Notes on putting AI to work.
Practical writing on corporate brains, AI agents, evals and model choice, from the team at Citrusworks.
Brains and operating systems
Build a personal AI OS that runs your week
A personal AI operating system is a set of skills, memory and scheduled agents built around one person's job. Here is what goes into one and how to start this week.
How to set up a team OS for AI agents
A team OS is the shared set of skills, memory, agents and scheduled loops a team runs its work on. Here is how to set one up in five steps.
What a corporate brain is, and what it is not
A corporate brain is one structured memory of what your company knows, readable by people and AI agents. What goes in it, and how to spot a fake.
Agents and evals
Five ways to use a voice profile
A free MCP server that measures how you write, with five practical uses: checking AI drafts, writing across channels, diagnosing your style, ghostwriting and house style.
Every agent needs a judge
An AI agent without evals is a demo. This is how to put a scoring gate in front of an agent so that only output that passes reaches a customer.
Models and classifiers
Pick the cheapest model that passes your evals
Frontier, open or fine-tuned? Choose the cheapest, fastest model that passes your evals. Here is how to run that decision.
How to build a classifier that holds up in production
Most AI work inside a company is classification. Here is how to choose between a prompt, a fine-tuned model and a classic model, and how to keep the result accurate.
When to train your own model, and how
Fine-tune your own model when a small model is cheap and fast enough but not accurate enough, and you have labeled examples. Here is the decision and the process.
Industry news
Jev vs OpenAI's Decisions API: models that decide instead of write
TypeSafe's Jev and OpenAI's Decisions API both return a typed decision with a probability instead of text. What each one is, what is still unproven, and what to do about it.