Introduction to Claude

Summary

Wei presented the fundamentals of Claude. The session covered how to choose a model for the task at hand, how token usage works, and the everyday habits that make Claude a reliable part of a workflow. It also showed how connectors let Claude reach external tools and files, and addressed common questions about large documents, memory, language preferences, and data handling.

On data, the presenters clarified that conversation data is held temporarily on the server, is not used to train the model, and follows professional data-handling practices.

Key takeaways

• The four-week course builds up from the basics to advanced features: models, prompting, artifacts and memory, projects, then skills and MCP.
• Match the model to the job: Opus for complex work, Sonnet for everyday use, Haiku for quick tasks.
• Tokens are the unit of work; longer context and heavier tools cost more.
• Connectors give Claude access to external tools and files.
• Your data is stored temporarily and is not used for training.
• Set memory and language preferences so Claude works the way you do.

Chapters

1. 00:00Programme introduction
2. 06:54Claude basics and best practices
3. 18:01Connectors demonstration
4. 23:18Workflow integration demonstration
5. 26:44Large document processing limitations
6. 32:17Best practices discussion and Q&A

Resources

• Recording, slides, transcript & chat 8 files on Zoom→