Welcome back

Today we finish the setup work from Day 1, then move from cleaning data to exploring it and producing our first forecasts.

The middle of the day is about prompting: how you ask shapes what you get back. By the afternoon you will be getting your first forecasts!

What today covers

By the end of today you will be able to:

  • Finish your VS Code, Copilot, and working directory setup
  • Produce a cleaned dataset and epicurve, then commit them and open a pull request
  • Explain what makes a prompt effective for AI assisted coding and analysis
  • Explore the cleaned data and generate summary visuals through your agent
  • Build and run a forecasting script, and interpret what it produces

Today’s flow

  1. Introduction to VS Code: R extension, GitHub Copilot, cloning your fork, and your working directory

  2. Activity 1: Data Cleaning and Visuals

  3. GitHub commits and pull requests

  4. Lecture 2 (Continued): Introduction to AI and Human-AI Teaming

  5. Lecture 3: Prompting for AI Assisted Coding and Analysis

  6. Activity 2: Data Exploration

  7. Lecture 1 (Continued): Overview of Forecasting

  8. Activity 3: Forecasting

Where we are in the pipeline

  • Activity 1: Data Cleaning and Visuals (finishing today)
  • Activity 2: Data Exploration (today)
  • Activity 3: Modeling and Forecasting (today)
  • Activity 4: Evaluation
  • Activity 5: Refinement

By the end of today you will have moved from raw CDC data to actual forecasts. Tomorrow is about judging whether those forecasts are any good and improving them.

Resources

Getting help

  • Ask a TA during activity blocks
  • Post in the course Slack channel
  • If your setup is still fighting you, flag it now rather than losing modeling time this afternoon