Policies

Generative AI Use Policy

This course is about configuring and directing generative AI coding agents, so using them is the central subject matter, not a gray area. What remains non-negotiable is transparency and demonstrated understanding.

Permitted and Expected AI Use (mini-projects and final project code)

• Writing, refactoring, and debugging code
• Planning and executing spec-driven development workflows
• Exploring concepts, generating examples, and drafting technical documentation
• Orchestrating subagents and automation as part of the graded harness itself

Restricted AI Use

• Weekly quizzes are closed-AI. No AI tools, devices, or notes during quizzes.
• Final Project poster prose must be human-written. AI may be used for brainstorming, outlining, feedback, and figure/code generation, and this must be noted in the Declaration of AI Use, but the text on the poster must be your own.
• Lightning presentations: AI may help you find and understand the paper. The summary and critique you present must be your own.

Required Disclosure

All mini-projects and the final project must include a Generative AI Use Statement in the README (or poster) describing which AI tools/agents were used and how, including any harness/hooks/subagents configured as part of the deliverable itself. This is expected and does not count against a project. For MP2–4 it is the project.

Evaluation Criteria

Work is evaluated on demonstrated understanding of how the agentic system was configured and why, not on whether AI was used to produce it. A mini-project that works but whose author cannot explain the harness design in the demo or in office hours will not receive full credit.

Academic Integrity

Submitting a harness or workflow you cannot explain, misrepresenting what you configured versus what came from an unmodified template or tutorial, or using AI in ways that prevent you from demonstrating the required learning outcomes constitutes academic dishonesty. If your project builds on a public tutorial or template repository, cite it explicitly and document your modifications and extensions.

Late/Missing Work Policy

• Late mini-project and final-project submissions are penalized 10% of the assignment grade per week (or portion thereof) past the deadline. Pass/No-Pass projects submitted late are treated as No-Pass until the resubmission window described in the Assignments page.
• Quizzes cannot be submitted late. Missed quizzes are covered by the two dropped scores and, beyond that, by documented excused absences.
• Extensions may be granted for documented emergencies if requested before the deadline.
• Missing work (not submitted) receives a grade of zero.
• No extensions are possible past the official end of the semester exam period (4:30 PM Friday, December 18, 2026).
• If you are struggling to meet a deadline, tell the instructor early. Earlier is easier for everyone.

Attendance Policy

Regular attendance is required. This is a seminar-style, project-based course where each session covers substantial and cumulative material.

• Attendance is part of the Participation grade. An unexcused absence results in an automatic 0 for that session's participation credit.
• Excused absences require documentation (illness, family emergency, college-recognized activities such as athletics, religious observance) and must be communicated to the instructor in advance where possible.
• Makeup work for excused absences will be arranged with the instructor.
• Religious observance: Students who anticipate a conflict between a religious holiday and a class session or deadline should notify the instructor within the first two weeks so accommodations can be arranged.

Student Work and External Sharing

Mini-project repositories may be public or privately shared with the instructor, at each student's discretion. All external sharing of student work beyond the instructor is voluntary and opt-in, with the exception of the Final Project, which is a public deliverable by design (or replaced by the written-exam alternative).

• Portfolio Development: Students are encouraged to make MP1–4 repositories public to showcase their work for graduate school or employment. Students control what they share publicly for MP1–4.
• Final Project — Digital Kenyon Publication: The final project poster is intended for publication at digital.kenyon.edu/dh as part of the course's public digital portfolio requirement. Students who prefer not to publish publicly may instead complete the 3-hour written exam alternative. A standard Kenyon Digital Release Form is required for publication.
• Research Collaboration: Selected mini-project work (e.g., benchmarking exercises) that passes quality-assurance review may be invited for co-authorship credit on an original research paper. Participation is voluntary.
• AI Training Data: In limited cases, anonymized portions of shared student work (with explicit written consent) may contribute to small labeled research datasets. Students will be informed and may decline without any effect on their grade.
• Default Setting: Opt-out is the default for all sharing beyond the Final Project's public-portfolio requirement (or its written-exam alternative). Students must explicitly consent to any further sharing.

Secrets Hygiene and Agent Safety

This course has you publish configuration repositories and give coding agents real permissions on your own machine, so the following are course rules:

• Never commit API keys, tokens, or passwords to any repository, public or private. Keep secrets in environment variables or in a .env file listed in .gitignore, and turn on GitHub secret scanning / push protection for your account. The first leaked key is a learning moment. After that it is a grading deduction.
• Run agents with least privilege. Do not run an agent in bypass-permissions ("YOLO") mode on a machine or directory holding data you care about. Use a dedicated project directory, rely on git for recovery, and (in MP3) write hooks that deny destructive commands.
• Keep confidential data out of agent context. Do not paste other people's private information, unpublished third-party research, or College-internal data into commercial AI services.
• Report any accidental leak to the instructor immediately so we can rotate keys. There is no penalty for prompt disclosure.

College Statements and Policies

Statement on Academic Honesty

At Kenyon College we expect all students, at all times, to submit work that represents our highest standards of academic integrity. It is the responsibility of each student to learn and practice the proper ways of documenting and acknowledging those whose ideas and words they have drawn upon (see Academic Honesty and Questions of Plagiarism in the Course Catalog). Ignorance and carelessness are not excuses for academic dishonesty.

In this course, proper attribution includes citing AI tools/agents and any third-party templates, tutorials, or FOSS frameworks used in completing assignments. Misrepresenting unmodified template/tutorial work as original, or submitting AI-generated work in ways that prevent demonstration of required learning outcomes, constitutes academic dishonesty.

Accessibility Accommodations

Kenyon College values diversity and recognizes disability as an aspect of diversity. Our shared goal is to create learning environments that are accessible, equitable, and inclusive. If you anticipate barriers related to the format, requirements, or assessments of this course, you are encouraged first to contact the office of Student Accessibility and Support Services (SASS) by emailing sass@kenyon.edu, then to meet with the instructor to discuss accommodation options or adaptations.

Title IX / Responsible Employee Statement

As a member of the Kenyon College faculty, I am concerned about the well-being and development of students, and am available to discuss any concerns. However, I want you to know that faculty members are legally obligated to share certain information with the College's Civil Rights & Title IX coordinator. This is to ensure the student's safety and welfare is being addressed. These disclosures include, but are not limited to: reports of discrimination or harassment due to a protected characteristic, sexual harassment, sexual assault, relational/domestic violence, and stalking. The Health and Counseling Center, the College chaplains, and the staff at New Directions Domestic Abuse Shelter & Rape Crisis Center are confidential resources.

Wellbeing

College is demanding and this course moves fast. If you are struggling, academically or otherwise, talk to me early. The Health and Counseling Center and your academic advisor are also there for you. It is much easier to make a plan in Week 6 than to dig out in Week 14.

Copyright of Course Materials

Course materials created by the faculty instructor such as slide presentations, handouts, assignments, outlines, quizzes, tests, and classroom recordings are protected by copyright law. You may share these materials with other students enrolled in the course. You may not reproduce, distribute, or display course materials for anyone outside of the class without the faculty member's explicit, written consent. Students are not permitted to record class sessions without the permission of the instructor.