AI Literacy
Stop treating every AI feature like the same magic trick. Learn what models can do, when they need your documents, and when a simple prompt beats building an agent. Free, with a lab and a quiz.
Start hereLearn · The Loaded Tokens desk
Understand AI. Then put it to work.
Start free. Go deeper with Pro.
New to AI? Start with the free AI Literacy course. Learn when to use a prompt, bring in documents, or reach for an agent, then try the lab and quiz. Ready to build? Create a small agent, add a tool, and test whether it runs only when it should. The architect course is independent prep for a real $125 exam: 120 original questions and a two-hour mock. Pro is listed at $19/mo and is not for sale yet.
Claude Certified Architect Decoded
Independent prep for Claude Certified Architect – Foundations. Knowing the terms is one thing. Choosing what an agent should do when a refund, a missing fact, or a bad tool call is on the line is another. Practice with domain lessons, worked examples, 120 original questions, and a two-hour mock.
Open the courseBuild Your First AI Agent
Build a small agent that can call a tool, save a note, and delete it again. Start with the included mock, so no API key is needed, then optionally connect a real model. Ten test cases help you check what happened.
Open the coursePrompting for real work
Next on the roadmap: prompts for the job you already have, with the boring constraints still in the prompt.
The AI bill
Next on the roadmap: how to read a vendor bill without inventing a usage number the invoice does not show.
Ship a feature past the demo
Next on the roadmap: the gap between a demo and a feature a coworker can open on Monday.
Step counts and minutes are the pages in each course, using the minute field on each page. Question counts are the questions in the bank. The comparison still shows the separate seat-time method, including the two-hour mock. The sample certificate is a learning milestone, not an accredited credential. The glossary has 54 terms.
Free explainers
Ten shorter pages, still free.
Diagrams, exercises, sources. They are not the courses above.
Start with what a language model does →UnderstandChooseBuild responsibly
Beginner
AI basics
Start here if the acronyms are piling up.
01 · 4 min read
AI is a prediction engine, not a tiny coworker
Understand the difference between producing plausible language and knowing that a claim is true.
Read the explainer →02 · 4 min readTokens, context and memory: three different tabs on the bill
Decode the units behind model limits without pretending a token is always a word.
Read the explainer →Beginner → intermediate
Models & how they differ
Pick a model for the job in front of you. The leaderboard can wait in the other tab.
01 · 4 min read
Choose a model like a vehicle, not a sports team
A practical scorecard for comparing capability, latency, data rules and cost.
Read the explainer →02 · 4 min readOpen weights are not a free lunch with a GPU garnish
Separate access to model files from licenses, hosting costs and operational ownership.
Read the explainer →Intermediate
Agents
How agents use tools and repeat steps, and what to check before you give one access.
01 · 4 min read
An agent is a loop. Your business still needs a brake.
Know when a fixed workflow beats a model choosing its own next action.
Read the explainer →02 · 4 min readMCP is the adapter. Permission is still the socket.
Understand tool interoperability without mistaking a connection for a safety guarantee.
Read the explainer →Intermediate → advanced
Building with AI
How a demo becomes something your coworker can open without calling you.
01 · 4 min read
RAG gives your model a reading list, not a truth serum
How retrieval-augmented generation connects evidence to an answer—and where it can still fail.
Read the explainer →02 · 4 min readYour AI demo needs a report card and a receipt
Evaluate quality and total cost per acceptable outcome instead of celebrating one lucky output.
Read the explainer →All levels
AI for work
Use AI to cut busywork while you keep making the calls.
01 · 4 min read
Stop writing magic spells. Write a useful brief.
A reusable prompt structure built around task, context, constraints and an inspectable output.
Read the explainer →02 · 4 min readThe workplace AI checklist: helpful, not accidentally public
Choose low-risk workflows, protect data and measure real effort saved without outsourcing accountability.
Read the explainer →