Fall 2026/Summer 2026 Course Bulletin

MATH 111, MATH 112 and COMPSCI 101 -- Only these three courses are accepted as S/U (if they can be taken S/U) and count towards degree requirements for CompSci majors and minors. For other majors and minors you plan to have, you need to check if they will also allow these courses to count if taken S/U.

For IDM majors, only these three courses -- MATH 111, MATH 112 and COMPSCI 101 -- are accepted as S/U (if they can be taken S/U) and count towards the CompSci part of your IDM. Check with your other major in your IDM to find out whether they will also allow these courses to count if taken S/U.

It is possible these courses may not be offered with the option to take them S/U.

1. You are now able to use COMPSCI 210D in lieu of the COMPSCI 250D requirement. If you have already taken COMPSCI 250D, you will not be able to take 210D. If you are an ECE major, you must take COMPSCI 250D.

2. You can now use COMPSCI 231D or COMPSCI 232 in lieu of the COMPSCI 230 requirement. If you are using any of the other substitutions in lieu of COMPSCI 230 or COMPSCI 231D or COMPSCI 232 as the prerequisite for a course--for example, MATH 371 and STA 230--you will need to request a permission number through the CS PIN request form. 

   NOTE: Dukehub does not recognize substitutions for prerequisites.

3. For the CompSci majors and minor, only one CompSci course can be included which falls into the category of Social and Policy Oriented Computing courses. Examples of Social and Policy Oriented Computing courses include COMPSCI 240, COMPSCI 247S, COMPSCI 255, COMPSCI 265S, COMPSCI 342 and COMPSCI 290/290D-001 - Cinema Perspectives on Artificial Intelligence.
4. For the CompSci BS and BA majors as well as the Software Systems and AI/Machine Learning concentrations, COMPSCI 345 - Graphics Software Architecture AND COMPSCI 512 - Distributed Systems now count as systems core classes.

COURSE SUBSTITUTION GUIDE

More About Special Topics Courses

Special Topics courses are new courses that we are trying out and may eventually give a permanent number.

COMPSCI 390-01 - Intro to Software Engineering | Instructor: Eric Fouh / This course will introduce students to the tools, processes, and techniques used by professional software engineers to create high-quality software, with a focus on software development and testing. Additionally, students will apply these in the creation of a software system. Prerequisite: COMPSCI 201.

COMPSCI 390-04 - AI-Copiloted Software Engineering Practicum | Instructor: Jun Yang and Alexander Steiger / Working in a software project team, students will develop a large, new software system, carrying the project through from requirements analysis to acceptance testing. AI-copiloting will be used throughout the process. Teams will make regular presentations on project planning, requirements, architecture, detailed design, quality assurance, and final product presentations, and reflections on the experience. Students will leave the course with a firsthand understanding of AI's impact on software development; they will have concrete experience with AI-copiloted development, and will have engaged in active reflection on this experience.

Prerequisites: COMPSCI  201, COMPSCI 210, and at least one CompSci course on systems and/or software development at 300-level or above. Prior experience with medium to large software development projects is expected.

COMPSCI 590-01 - Software Design & Engineering | Instructor: Eric Fouh / This course will familiarize students to the tools, processes, and techniques used by professional software engineers to design, create and deploy high-quality software, with a focus on software design, development and testing. Additionally, students will apply these in the creation of a software system. Recommended prerequisites for grad students: 1 year of programming.

COMPSCI 590-02 - Building Intelligent Agents with Frontier Models | Instructor: Shuyan Zhou / Graduate-seminar course. This course explores the foundations and recent progress in the design and development of AI agents powered by Large Language Models (LLMs). A central theme is the transformation of language models from passive generators into autonomous agents capable of pursuing goals, reasoning over time, and interacting with complex, dynamic environments. The curriculum covers core methodologies such as pretraining, scaling laws, and reinforcement learning, alongside specialized topics including tool use, computer and browser interaction, software engineering agents, and embodied control. Students will investigate advanced paradigms like multi-agent systems, world models, and self-improving agents while also addressing benchmarking, evaluation, safety, and the societal impact of automation. The course is delivered through a blend of lectures, student-led presentations, and a deep-dive research project. Cross ECE 590-02.

Recommended prerequisite: For undergraduate students: (COMPSCI 371 or COMPSCI 671) and (COMPSCI 572 or COMPSCI 574) For graduate students: COMPSCI 671 and (COMPSCI 572 or COMPSCI 574); or Linear algebra; Machine Learning; Deep Learning / Natural Language Processing; Calculus; Python.

COMPSCI 590-04 - Robot Studio | Instructor: Boyuan Chen / This is a hands-on studio course that will expose students to the entire robot design process “from A to Z”, including kinematics, industrial design, manufacturing, electronics, simulation, algorithms, and programming. The course project can be done individually or in pairs. The goal of the course is to design, build, and control an organic-looking four-legged robot, using up to twelve motors in total. Cross ME 555-04.

COMPSCI 375 - Natural Language Processing | Instructor: Dhingra / This new course will count as a general elective for the CompSci BS and AB degree requirements or as an elective for the AI/ML concentration. This is an introductory course in natural language processing, covering core computational problems involving human language, and both classical techniques such as bag-of-words, n-grams and and conditional random fields, as well as modern neural approaches such as word vectors, recurrent neural networks and transformers. We will also lay the foundation for understanding large language models and their scaling properties. Focus will be on key NLP tasks such as text classification, language modeling, machine translation and information retrieval. Prerequisite: COMPSCI 201 and familiarity with Python. It is recommended, though not required, to also have completed one of COMPSCI 270 or COMPSCI 371 or COMPSCI 372.

Most of these new courses have been taught previously as Special Topics courses (i.e., COMPSCI 290, 390, and/or 590) and now have permanent course numbers. If you have already taken them as a Special Topics course, you may NOT take the re-numbered course.