A short, visual explanation of how a model turns your words into an answer — no maths, no code. Watch this before you decide whether to trust a particular answer, because most over-trust comes from not knowing what the thing is doing. You'll finish with a working mental model you can explain to a colleague in a minute.
A repeatable structure — role, task, context, format — that turns a vague request into a usable draft. The example is ordinary office work, not a clever demo, so you can copy the shape straight onto your next task. If your answers have been generic, this is usually the reason.
A plain walkthrough of the interface: conversations, files, custom instructions, and where your settings actually live. It's aimed at someone who has typed a question into the box once and never looked at the rest. Watch it once and stop guessing which button does what.
A beginner's tour of Claude and how working with it differs in practice — long documents, longer context, a different feel to the writing. Useful when you're deciding which assistant to open for a given task rather than defaulting to one. No comparison-chart hype, just what each is comfortable doing.
How AI meeting notes work in a real tool: what gets captured, what gets summarised, and what you still have to write yourself. The decisions and owners are the part machines are worst at, so you'll see where to check before you send. Worth watching if the write-up is the part of every meeting you keep putting off.
A worked example of drafting outreach with a model, including the research you have to feed it for the result to be specific to the recipient. The lesson underneath it is the one that matters: generic input produces generic email, and everyone can tell. You'll leave with a drafting process, not a template to paste.