Course Launch and Working Effectively with AI

Monday established why the course is changing

• AI has made producing code, prose, and prototypes dramatically cheaper.
• Polished artifacts no longer reliably reveal effort, learning, or judgment.
• The valuable skill moves upstream: principles, technical intuition, discernment, and value.
• Methods remain important—but as vehicles for understanding durable ideas.

The redesign separates learning from effective doing

Several details are now clearer than they were Monday

• Oral quizzes now have a defined multi-question, retake, and support flow.
• AI Products remain individual and standalone even when students share stakeholders.
• The real- and virtual-stakeholder paths have the same course-grade eligibility.
• Dynamic topic decks and frozen class-day slides are both available on the course website.
• Course material stated or explained in lecture—even when it does not appear on a slide—is fair game for quizzes and exams.

Open study groups need one shared directory and one contact

• Use one editable Piazza post as the class study-group directory.
• Every listed group must welcome any student in the course.
• Each listing names:
  • recurring time, place or online link, and focus;
  • one designated coordinator and Purdue email; and
  • whether the meeting details have changed.
• The coordinator keeps a contact list of everyone who has attended.
• If the time, place, or link changes, the coordinator notifies everyone on that list before the meeting.
• The coordinator manages logistics—not membership or access.

OralScale makes quizzes repeatable practice

• Open the release post on Piazza. Gradescope controls due dates and submission.
• Quiz 0: ungraded practice, five questions, one continuous recording.
• Graded quizzes: five questions, one continuous recording.
• Retakes draw fresh questions.
• Complete an attempt to retake immediately. Abandoned attempts wait about five minutes.

The AI Product now has clearer paths and submission support

• Build a small, credible product whose primary use case materially depends on AI.
• Create and test a stakeholder value claim; complexity and feature count are not the goal.
• Choose either the encouraged real-stakeholder path or the full-credit virtual path.
• Shared stakeholders are allowed; every student’s product and evidence remain separate, and grades remain independent.
• Milestone packages provide the exact structure, agent guidance, checker, and packager.

The AI Product documents are the current authority

What questions or concerns do you have about the AI Product?

• Finding or sharing a stakeholder?
• Choosing between the real and virtual paths?
• Scoping something meaningful but feasible?
• Understanding evidence, grading, privacy, or submission format?

We will finish the syllabus and schedule walkthrough

The syllabus is required reading. Quiz questions may assess syllabus content.