Course Outline
Phase 1 — Getting Started with Claude Code — 30 minutes
- Overview of Claude and the unique capabilities of Claude Code compared to standard chat interactions
- Orientation session: Focus on the Claude app (web or desktop); note that CLI and other interfaces are detailed in the reference materials
- Navigating the interface: Initiating coding sessions and familiarizing yourself with the workspace
- The internal logic of Claude Code: Understanding the Describe → Plan → Act → Review cycle
- Permission management: Explaining why Claude seeks confirmation before generating files or executing code
- Initial build challenge: Instructing Claude to generate a simple styled webpage based on a single-sentence prompt
- Refining outcomes: Adjusting results with instructions such as “increase header size,” “modify color palette,” or “insert navigation bar”
- Practical activity: Learners initiate a session and construct a custom “About Me” webpage, iterating through follow-up directions
Objective: Ensure all participants overcome the initial usage barrier and feel at ease with the interface.
Intermission — 7 minutes
Phase 2 — Creating Functional Projects via Natural Language — 55 minutes
Completion of three progressively complex tasks using only natural language prompts.
- Exercise 1 — Interactive Dashboard: Constructing a styled dashboard featuring sample data, charts, and statistics. Practicing design directives like “apply dark theme,” “insert sidebar,” or “ensure responsiveness.”
- Exercise 2 — Data Analysis: Providing a sample CSV file for Claude to summarize, identify trends, detect highs/lows, and generate visual charts. This demonstrates Claude’s ability to write and execute code on the user's behalf.
- Exercise 3 — Automation Utility: Building a simple tool, such as a unit converter, quiz application, or budget calculator. This highlights Claude’s capacity to create interactive tools rather than just static pages.
Following each task, the instructor reviews the backend actions: files generated, code written, and output interpretation. Participants record their most effective prompts in a shared Prompt Playbook.
Intermission — 7 minutes
Phase 3 — Optimizing Workflow with Claude Code — 35 minutes
- The craft of effective prompting: Distinguishing between specific and ambiguous instructions
- Live demonstration: Side-by-side analysis of weak versus strong prompts applied to the same objective
- Refinement techniques: Requesting explanations of choices, undoing changes, or exploring alternative approaches
- Processing uploaded files: Tasks such as “read this document for a summary” or “convert this spreadsheet into a chart”
- Multi-step processes: Chaining requests to generate complex outcomes
- Resource management: Understanding tokens, context windows, and subscription tiers
- Strategic usage: Determining when to utilize Claude Code versus standard Claude chat
- Practical activity: Participants enhance a Phase 2 project with a new feature using multi-step prompts, then analyze before-and-after prompts to pinpoint key differences
Objective: Transition from basic functionality to consistently achieving high-quality results.
Intermission — 7 minutes
Phase 4 — Integrating Claude with External Tools via MCP — 34 minutes
Preparation: Participants received instructions prior to the session to connect Gmail or Google Drive, allowing classroom time to focus on application rather than authentication.
- MCP (Model Context Protocol) overview: The universal integration standard for AI tools
- The value of MCP: Evolving Claude from a chat assistant into a connected workflow hub
- Connectors Directory: Discovering and adding integrations directly within the Claude app
- Desktop Extensions: One-click installation options (available for Claude Desktop users)
- Live demonstration (single workflow): “Review my Google Calendar for tomorrow’s meetings and draft a preparation email for each”
- Practical activity: Participants utilize their pre-configured service (or connect one live) to assign a task to Claude — e.g., “Review recent emails regarding project updates and generate a summary document”
- Key concepts: OAuth, permission management, controlling tool access per conversation, security considerations, and locating new connectors
Objective: View Claude as a connective layer rather than merely a coding tool.
Phase 5 — Capstone & Conclusion — 35 minutes
Capstone Mini-Project (25 min): Each participant selects one of the following scenarios:
- A refined landing page or portfolio website
- A data analysis pipeline: Uploading a file, analyzing it, and producing a visual report
- An interactive tool that solves a specific workflow problem
- A connected workflow: Extracting data from the Phase 4 service, transforming it, and generating a deliverable
The instructor provides individual assistance, helps refine prompts, and highlights exemplary work.
Conclusion (10 min):
- Next steps: Claude Code CLI for terminal users, VS Code extension for developers, and Cowork for knowledge workers
- Subscription plans: Free vs. Pro vs. Max — detailing features and ideal use cases for each
- Recommended resources: Official documentation, Anthropic’s prompt engineering guide, and community forums
- Takeaways: Participants receive a reference card summarizing prompting patterns, connector setup, and essential MCP integrations
Requirements
Prerequisites
Knowledge Base
- Basic computer skills: Managing files and folders, using web browsers, and installing software
- Familiarity with AI assistants: Casual use of ChatGPT, Gemini, or Claude is beneficial but not mandatory
Experience Level
- No coding, programming, or terminal experience is necessary. This course is tailored for individuals who have never written code.
- No prior experience with Claude or other AI tools is required.
Technical Specifications
- Participants must bring a laptop (Mac, Windows, or Linux) equipped with a current web browser
- A stable internet connection
- A Claude Pro subscription for the duration of the session (included in course registration; setup guidelines are provided before class)
- Claude Desktop is suggested but not mandatory (the claude.ai web app supports all exercises)
- A Google account is recommended for the MCP connector exercise (Gmail, Google Drive, Google Calendar), though alternative connector options exist
Target Audience
- Business professionals seeking to enhance productivity and automation through AI
- Marketers, operations managers, and analysts aiming to automate repetitive tasks
- Founders and entrepreneurs looking to build prototypes without hiring developers
- Educators and researchers exploring AI-assisted workflows
- Individuals without technical backgrounds who are curious about Claude’s capabilities
Testimonials (1)
That i gained a knowledge regarding streamlit library from python and for sure i'll try to use it to improve applications in my team which are made in R shiny
