2 Your side of the conversation

Claude is like a real human assistant — the more you work together, the more effective you are. The more you use Claude, the better you understand how to get the best out of your AI assistant, and the more comfortable you become handing work over. You gain confidence in Claude’s capabilities, and you know where it will save you time and improve your delivery. You should soon get to the point where Claude is your first point of contact for every task — you will either delegate it entirely to Claude, or use Claude to research and refine a plan.

To get there you need to spend time working with Claude on different types of task, understanding how the platform works and exploring the boundaries. There are two parts to that - knowing the practical features of the tool so you can use it productively, and understanding how your prompts affect the conversation and direct the output. These are the techniques you build up to write clear requests, manage your sessions and artifacts, and evaluate and steer responses when they’re not right.

2.1 How Claude works on web and mobile

2.2 Building your register

2.3 Claude’s knowledge and limits

2.4 Starting with a good first prompt

2.5 Working practically with Claude

2.5.1 Managing your conversations

2.5.2 Managing Claude’s output

2.6 Structured and unstructured data

2.6.1 Try it now — generating RFCs

2.6.2 Reading and synthesizing unstructured inputs

2.6.3 Try it now — assess a technical approach

2.6.4 Try it now — build a design system

2.7 Evaluating, iterating, and recovering

2.7.1 When Claude gets it wrong

2.7.2 Pushing back effectively

2.8 Your turn: produce the component designs for your app

2.9 Summary