Sprint Project: Dashboard from Scratch

Your mission

You have 60 minutes to build a working interactive dashboard from a raw dataset using only LLM prompts. No manual coding allowed. Then present your approach to the class.

The Challenge

Build an interactive web dashboard (Streamlit, Shiny, or similar) from scratch without typing code. The dashboard must load data, display at least two visualizations, and include at least one interactive control. You are forbidden from manually editing code. If the LLM generates errors, prompt it to fix them.

Kickstarter Prompts

Example prompts to get started:

  1. “Create a Streamlit app that loads data.csv and displays the first 5 rows in a table”
  2. “Add a sidebar with a dropdown to select a column from the dataset”
  3. “Add a scatter plot showing the relationship between column X and column Y”
  4. “Make the scatter plot update when the user selects different columns from the dropdown”

Remember: You cannot edit the generated code manually. If there are errors, describe them to the LLM and ask it to fix them.

The Rules

Time: 60 minutes of work, followed by presentations.

No Manual Coding: All code must come from LLM outputs. You can copy-paste but not edit by hand.

Tool Choice: Use any LLM assistant (ChatGPT, Claude, Copilot). Choose any framework (Streamlit, Shiny, etc.).

Working Demo: App must run locally without errors and respond to user interactions.

Evaluation

Speed (40%): First working demo wins. Basic functionality beats ambitious failures.

Quality (40%): If multiple teams finish, judges evaluate clarity of layout, labels, and interactivity.

Presentation (20%): Explain your prompting strategy and key challenges overcome.

Deliverables

Your submission should include:

  1. Dashboard Code: Complete working application code
  2. Prompting Log: Documentation of key prompts you used and their effectiveness
  3. Report: A brief markdown document explaining:
    • Your dashboard design strategy
    • Prompting approach and lessons learned
    • Challenges with LLM-generated code and how you resolved them

Submission

  1. Use the provided template: https://github.com/sk-classroom/sprint-project-template
  2. Follow the template instructions to create your project repository
  3. Place your dashboard code in the workflow folder
  4. Document your prompts in README.md
  5. Include screenshots of your working dashboard
  6. Submit the link to your GitHub repository to Brightspace