Syllabus
Common Questions
- Do I need to pay for Claude Code? Yes. Students are required to purchase their own Claude Code subscription at the Pro level ($20/month), for the duration of the course, as a course cost. You can purchase a subscription at claude.com/pricing. If this creates a financial hardship, contact course staff as soon as possible to discuss your situation.
- Am I allowed to use AI on this course? Yes — it's required. See the AI Use Policy below; the short version is that using Claude Code is expected on every project, but you must be able to explain, defend, and verify anything it produces.
- Quizzes: We do not offer extensions or reopen closed quizzes because we release the answers. We drop your 5 lowest quiz scores and your 5 lowest lecture-activity scores to account for illness, technical difficulties, or other unforeseen circumstances.
- Slip days: Each student has a bank of 5 slip days to use across P1–P4 (not the Final Project). Slip days waive the late deduction but never move the hard cutoff — see Late Policy below.
- Manual Grading: Project grading is results-based (does your code pass the auto-grader / meet the rubric) combined with a review of your git commit history to confirm both partners meaningfully contributed.
- Getting Help: Office hours and email. Office hours are the fastest way to get unstuck. See Tools → Scheduler (scheduler.hub.cs.wisc.edu (opens in new tab)) for this week's office hours times and location.
- I missed lecture: Check the schedule page for that lecture's pre-class readings and in-class activity. Do the reading, attend office hours with questions, and complete the activity before the next lecture.
Course Description
The software engineering landscape is shifting rapidly, and this course is designed to ensure you're leading the charge. In AI-Assisted Software Development and Intelligent Applications, you will master the two most critical skills for the modern developer: using AI to supercharge your own productivity and engineering the next generation of AI-powered software. We will dive deep into full-stack development, showing you how to leverage AI tools across the entire software life cycle — from initial system design and rapid coding to automated testing and cloud deployment — while teaching you how to architect and integrate intelligent models directly into your own applications. By the end of the term, you won't just be a faster coder; you will be a high-level engineer capable of building complex, AI-native systems from the ground up.
Official Course Description: This course examines the dual role of artificial intelligence in modern software development: as a transformative tool for developer productivity and as a core architectural component of next-generation software. Students will master AI-assisted development workflows to optimize the entire software development life cycle, leveraging intelligent tools for requirements analysis, code generation, debugging, and automated testing. The course provides rigorous, hands-on experience in full-stack development, teaching students how to architect and integrate large language models (LLMs) and AI agents into scalable applications. By utilizing industry-standard tools for version control, CI/CD, and cloud deployment, students will gain a comprehensive understanding of building, managing, and maintaining AI-native software at every stage of its life cycle.
Additions To Syllabus Made During Semester
- None yet
Course Learning Outcomes
By the end of the course, students should be able to:
- Leverage AI-Assisted Tooling: Integrate AI-driven development tools (such as Claude Code or similar agents) to accelerate code generation, refactoring, and documentation throughout the software life cycle. Use AI agents to manage version control software, containerization, and deployment tools.
- Architect Intelligent Systems: Design and implement full-stack applications that incorporate Large Language Models (LLMs) and AI agents as core functional components rather than simple external features.
- Implement Advanced Retrieval Patterns: Build and optimize Retrieval-Augmented Generation (RAG) pipelines using vector databases to provide applications with domain-specific context and memory.
- Manage the Full-Stack Life Cycle: Execute the end-to-end development of a software product — from initial system design and API integration to automated testing and cloud deployment.
- Evaluate AI Quality & Security: Critically assess AI-generated code and model outputs for technical accuracy, security vulnerabilities, and ethical considerations (such as bias and "hallucinations").
- Automate Software QA: Utilize AI to generate comprehensive unit tests, integration tests, and performance benchmarks to ensure the reliability of complex, non-deterministic applications.
- Optimize AI Performance: Balance technical trade-offs regarding model selection, latency, API costs, and token usage to create production-ready, scalable software.
Course Instructors
- Louis Oliphant (Teaching Faculty - Department of Computer Sciences) ltoliphant@wisc.edu
- Kaiser Pister (Teaching Faculty - Department of Computer Sciences) kaiser@cs.wisc.edu
Prerequisites
CS 200 and CS 300 required. Students should have taken or be concurrently enrolled in CS 400.
Course Credits
- 3 credits
Lectures (Meeting Time and Location)
- LEC003 Noland 132, Tuesday and Thursday, 1:00 PM - 2:15 PM
- LEC006 Morgridge 2522, Monday and Wednesday, 4:00 PM - 5:15 PM
Lecture recordings will not be provided. In-person attendance is expected.
Instructional Modality
- In-person
Computer & Tool Expectations
- Students need a laptop (Mac or Windows) capable of running VS Code, Node.js, Python, and Docker Desktop.
- Students are required to purchase their own Claude Code subscription at the Pro level ($20/month) for the duration of the course, available at claude.com/pricing. Contact course staff as soon as possible if this cost creates a financial hardship.
- Students will need to deploy applications to Google Cloud Platform (GCP); educational credits will be provided for course-related usage.
Communication Tools
- Canvas: Grades and announcements are posted on Canvas.
- Email: Please attend office hours with questions before emailing instructors.
- Office Hours:
- Louis Oliphant: Tuesday 2:30–3:30 PM, Wednesday 10–11 AM, Thursday 11 AM–12 PM, all in Morgridge Hall 5538.
- Kaiser Pister: After class, Monday and Wednesday, in Morgridge Hall 5534.
- Teaching Assistants: held at the tables just outside Morgridge Hall room 5538. TA hours vary week to week — see Tools → Scheduler (scheduler.hub.cs.wisc.edu (opens in new tab)) for this week's schedule and location.
Grading Breakdown
Your overall grade is based on the following:
- Quizzes: 7% (about 27 quizzes, one per lecture; the 5 lowest quiz scores are dropped)
- Surveys: 1% (several surveys throughout the semester asking how the course is going, used to improve the course for future semesters)
- Projects: 54% distributed as follows:
- P1 — Frontend Side-Scroller Game: 8%
- P2 — Full-Stack Quiz Game: 11%
- P3 — RAG-Based Smart Q&A Application: 11%
- P4 — Course Copilot (Local Multi-Agent Study Assistant): 11%
- Final Project — Full-Stack Smart Application: 13%
- 3 Exams: 30% distributed as follows:
- Midterm 1 (Unit 1 content): 10%
- Midterm 2 (Unit 2 content): 10%
- Unit 3 Exam (Unit 3 content): 10%
- In-Class Participation: 8% (attendance and engagement)
- Lecture Activities: most lectures include a hands-on in-class activity; the 5 lowest scores are dropped.
Graded Component Details
Quizzes:- A short quiz during every lecture testing pre-class reading comprehension.
- Out of 27 quizzes, your 5 lowest scores are dropped. This is meant to accommodate emergencies, technical difficulties, or other unforeseen circumstances.
- Extensions are not given on quizzes because we release the answers when the quiz closes.
- Most lectures have a hands-on in-class activity building toward that unit's connected project arc.
- Activities use a parallel practice codebase — never your project repo — so you can focus on the concept without project pressure.
- Your 5 lowest lecture-activity scores are dropped.
- Grading is batched, not immediate: lecture activities are graded twice a week. Activities from Wednesday/Thursday lectures are graded Tuesday night of the following week; activities from Monday/Tuesday lectures are graded Thursday night of that same week. Don’t expect your score right after submitting — it can take a few days to post.
- Projects primarily focus on your ability to design, build, and understand AI-assisted software.
- Please see the course schedule for project due dates.
- Teams:
- P1–P4 are completed in instructor-assigned pairs.
- The Final Project is completed in self-chosen teams of up to 4 students.
- Both partners are expected to meaningfully contribute. We review GitLab commit history per partner as the primary signal; if the commit history is too lopsided, the entire team will be deducted points.
- Submission:
- Each project is its own standalone GitLab repo. Clone, branch, implement, push, and open a merge request; the CI/CD pipeline auto-grades your main branch.
- You may push as many times as you like before the deadline, but the grading script will only run at designated times (e.g. right after the due date).
- Late Policy — Slip Days:
- Each student has a bank of 5 slip days for the semester, usable on P1–P4 only (not the Final Project).
- While you have slip days remaining, using one waives the late deduction for that day.
- Once your slip day bank is exhausted, each additional late day costs a 10 percentage point deduction.
- Hard cutoff: no submission is accepted more than 2 days after the due date, for any project, regardless of slip days remaining. Slip days waive the deduction within that 2-day window — they do not extend the window itself.
- The Final Project has no slip days and no exceptions to its due date.
- Project Grading:
- Grading combines automated unit tests (including hidden tests), LLM-graded rubric criteria, a review of AI logs, and a review of commit history for balanced partner contribution.
- For each project there is a required meeting with TAs for them to question you about your codebase and for any adjustments to errors in the automated portion of the grading.
- Each exam covers primarily the content from its unit.
- Midterms 1 and 2 are traditional written exams, closed-book, closed-laptop. The Unit 3 exam is an online Canvas quiz proctored with Honorlock.
- Students may bring 1 sheet of notes (8.5" x 11").
- Please see the course schedule for exam dates. Locations: TBD.
AI Use Policy
This course takes the opposite stance from most CS courses: Claude Code is a required, expected tool on every project. You are not just permitted to use AI — you are expected to use it as your primary development driver, the same way you'd use a compiler or a debugger.
The line we care about isn't "did you use AI." It's: can you explain, defend, and verify what it produced? This is the whole point of the Elephant-Goldfish Model workflow taught starting in Lecture 4 — design docs written before any code, a "Goldfish test" where a fresh AI session with no context has to be able to understand your design from your documentation alone, and a "mean code review" where you direct AI to tear apart your own finished code looking for problems. These exist so you demonstrate real understanding of what gets built, not just submit whatever the model outputs first.
Acceptable: using Claude Code (or another AI tool) to design, implement, debug, test, and review your project code, provided you can explain how it works, why it was built that way, and what its limitations are if asked in office hours or during a code review; and logging your AI sessions as required for each project.
NOT Acceptable: submitting AI-generated code you cannot explain, defend, or reproduce the reasoning behind; using AI on quizzes or written exams (which are closed-AI, closed-book); having a non-partner (human or AI acting on their behalf) do your share of a paired project's work; failing to log your AI sessions as required.
Using AI in a way that violates this policy is academic dishonesty, even though AI usage itself is expected and required in this course.
Academic Integrity
Standard UW-Madison academic integrity policy applies. The AI Use Policy above supplements it for AI-specific scenarios in this course.
Letter Grades
At the end of the semester, we will assign final grades based on these thresholds:
- 93% - 100%: A
- 88% - 92.99%: AB
- 80% - 87.99%: B
- 75% - 79.99%: BC
- 70% - 74.99%: C
- 60% - 69.99% D
Letter grade ranges include decimal points; we round scores to the hundredths place at the end of the semester.
No extra credit is given in this course.
How to Succeed in This Course
The best way to succeed in this course would be to:- attend in-person lectures with classroom engagement
- complete the pre-class reading before each lecture — the quiz and activity both depend on it
- treat AI-generated code as a first draft you must understand, not a final answer
- start projects early; slip days run out fast across four projects
- attend office hours when your auto-grader tests fail rather than guessing
Accommodations
Please notify the instructor within the first two weeks of classes:
- If you participate in religious observances that may conflict with course requirements.
- If you have received accommodations from the McBurney Disability Resource Center.
- If you have any trouble accessing or using the technologies used in this course (such as Canvas, GitLab, or the course website).
- If you need special accommodations for exams or other components of the course. We'll do our best to accommodate you.
Please notify your instructor if you suffer from an extended (> 12 days) illness.
Student Grievance Procedure
Any student at UW–Madison who feels that they have been treated unfairly has the right to voice a complaint and receive a prompt hearing of the grievance. The basis for a grievance can range from something as subtle as miscommunication to the extreme of harassment.
Ensuring that all students feel welcome and supported is a top priority of our department. The Department of Computer Sciences has developed procedures to handle complaints or incidents that are student-related, such as bias or mistreatment, or academic, such as a grade dispute or incident with an instructor. Information on the CS reporting procedures can be found here: https://www.cs.wisc.edu/report-an-incident/.
For assistance in determining options, students can also contact the drop-in staff member within the Dean of Students Office at 608-263-5700, within Bascom Hall, Room 70, Monday–Friday, 8:30 a.m.–4 p.m.
Official Statements Required on the Syllabus
Requisites
CS 200 and CS 300 required. Students may be concurrently enrolled in CS 400, or may have already completed it.
Regular and Substantive Student-Instructor Interaction: Substantive interaction is engaging students in teaching, learning and assessment through at least two of the following: direct instruction, providing feedback on student work, providing information about course content, facilitating discussion of course content, or other substantive interaction. Regular interaction is: predictable and scheduled interaction with students consistent with the course length (usually at least weekly but more often in a course of short duration). Regular and substantive student-instructor interaction, as defined by the US Department of Education (Within 34 C.F.R. §600.2), is always a requirement of UW-Madison for-credit learning activities.
Course Evaluations: UW-Madison now uses an online course evaluation survey tool. In most instances, you will receive an official email two weeks prior to the end of the semester when your course evaluation is available. You will receive a link to log into the course evaluation with your NetID where you can complete the evaluation and submit it, anonymously. Your participation is an integral component of this course, and your feedback is important to me. I strongly encourage you to participate in the course evaluation.
Academic Integrity: By virtue of enrollment, each student agrees to uphold the high academic standards of the University of Wisconsin-Madison; academic misconduct is behavior that negatively impacts the integrity of the institution. Cheating, fabrication, plagiarism, unauthorized collaboration, and helping others commit these previously listed acts are examples of misconduct which may result in disciplinary action. Examples of disciplinary action include, but is not limited to, failure on the assignment/course, written reprimand, disciplinary probation, suspension, or expulsion.
Accommodations for Students with Disabilities: The University of Wisconsin-Madison supports the right of all enrolled students to a full and equal educational opportunity. The Americans with Disabilities Act (ADA), Wisconsin State Statute (36.12), and UW-Madison policy (Faculty Document 1071) require that students with disabilities be reasonably accommodated in instruction and campus life. Reasonable accommodations for students with disabilities is a shared faculty and student responsibility. Students are expected to inform faculty [instructors] of their need for instructional accommodations by the end of the third week of the semester, or as soon as possible after a disability has been incurred or recognized. Faculty will work either directly with the student [you] or in coordination with the McBurney Center to identify and provide reasonable instructional accommodations. Disability information, including instructional accommodations as part of a student's educational record, is confidential and protected under FERPA.
Diversity and Inclusion: Diversity is a source of strength, creativity, and innovation for UW-Madison. We value the contributions of each person and respect the profound ways their identity, culture, background, experience, status, abilities, and opinion enrich the university community. We commit ourselves to the pursuit of excellence in teaching, research, outreach, and diversity as inextricably linked goals. The University of Wisconsin-Madison fulfills its public mission by creating a welcoming and inclusive community for people from every background – people who as students, faculty, and staff serve Wisconsin and the world.