CMSC 5443 AI Software Systems Integration

	Role	Office hours	Office location	E-mail address	Phone
Jicheng Fu	Professor	Monday, Wednesday, and Friday: 1:00 – 1:40 pm	STEM237	jfu@uco.edu	974-5704

Lecture Time & Location: Monday, Wednesday, and Friday 10:00 am - 10:50 am, MCS113

Course Web Page: https://cs2.uco.edu/~fu/SE4443/graduateindex.html

Getting Help

General questions about the homework assignments should be directed to the instructor at the above e-mail address. You are encouraged to use the eLearning discussion group for the course on D2L. Therefore, you should check the discussion group whenever you have a question about the assignment as someone else may have already asked it and received an answer.

Course Description

This course offers an advanced exploration of the integration of cutting-edge AI technologies into modern software systems. Students will engage in in-depth analysis and hands-on experimentation with generative AI frameworks, sophisticated API design and integration, autonomous agent architectures, and techniques for enhancing models with external knowledge across text, image, audio, and video modalities. Students will develop the skills necessary to design, implement, and manage scalable, secure, and ethically grounded AI-driven applications.

Suggested Text

Rush Shahani, “Building Reliable AI Systems”, Manning, ISBN 9781633436732

Objective

Upon successful completion of this course, students will be able to:

(1) Analyze and critique the fundamental components and advanced design patterns underlying modern AI software integration, integrating theoretical insights with practical application.

(2) Design, develop, and refine sophisticated API interfaces and AI-enhanced multimodal applications using cutting-edge open-source tools and industry-standard frameworks, while exploring novel integration strategies.

(3) Implement and optimize scalable, secure, and high-performance deployment solutions, ensuring robust operational efficacy in dynamic environments.

(4) Apply rigorous testing, benchmarking, and monitoring methodologies to assess the performance, reliability, and user experience of AI-driven systems, and use empirical insights to drive iterative improvements.

(5) Critically assess the ethical, legal, and societal dimensions of AI integration, and propose innovative, responsible solutions.

Prerequisite

SE 4283/CMSC 5283 Software Engineering I

Grading

Homework assignments (30% of the course grade) 
There will be a written as well as a programming assignment for each topic. Since we will be using eLearning (D2L) for submission and grading, you must upload an electronic copy of your assignment by the due date. For written assignments, if you choose not to typewrite your assignment, you will need to scan and upload your submission.

Research-based library assignment (5% of the course grade)

Using UCO library and database search facilities, locate, access, and read one article published within five years by the Association for Computing Machinery (ACM) or by the Institute of Electrical and Electronics Engineers (IEEE).  The article must be relevant to topics of AI Software Systems Integration.

Midterm exam (30% of the course grade) 
There will be one in-class midterm exam. The midterm exam is tentatively scheduled on Wednesday, October 21.

Final exam (30% of the course grade) 
There will be a comprehensive final exam. Exam date: December 11.

Class participation (5% of the course grade)

· A student is allowed to miss two complete classes without penalty. After that, unexcused absence will be counted. 

· For IVE students:

o Please use your real name to join our online sessions

o If you cannot make the class, you must watch the course video within 48 hours after the class. Please update your attendance records in D2L accordingly.

· To improve the learning quality, students are encouraged to actively ask questions, answer questions, and get involved in discussions. Attitude is everything.

Course Policies

Collaboration policy

The written assignments are individual and programming assignments are either individual- or group-based. Each group consists of at most two students.

Academic integrity policy

You are expected to maintain the utmost level of academic integrity in the course, in accordance with the academic integrity policy of the University of Central Oklahoma. In particular, (a) it is your responsibility to protect your work from unauthorized access, and (b) the work you submit is expected to be your own. Academic dishonesty has no place in a university or anywhere else: it wastes our time and yours, and it is unfair to everyone else. Any violation of this code will be penalized, as we take this issue very seriously. Any student observed cheating will receive a grade of zero on the exam or assignment, and the appropriate college administrative personnel contacted. A second offense will result in dismissal from the class with a grade of F.

Late assignment policy

Barring extenuating circumstances, all assignments must be turned in on the date specified. You will be given three ''free'' late days, with the restriction that no more than two free late days can be spent on each homework assignment. If it is a group assignment, students of the entire group will be considered using the free late days. After you use up the free late days, the late submissions will be penalized as follows. Assignments turned in within 24 hours of the due date will receive 90% of its score. Assignments turned in within 48 hours of the due date will receive 70% of its score. Assignments more than 48 hours late will not be accepted.

Regrade policy

The professor will grade your work carefully. However, questions about grading do occasionally arise. If so, first read the solutions. If questions persist, please see me of that problem (come to office hours or schedule an appointment). In the interests of smooth administration and to encourage you to look at your graded work soon after it is returned, regrade requests must be made within two weeks of when the work was returned. We reserve the rights to make regrade decisions "off-line" (i.e., not immediately at the time requested).

COURSE VIDEOS

Due to limitations on the disclosure of personally identifiable information under certain federal privacy laws, students are not permitted to record class sessions or allow non-students to view online class sessions. Sharing links of class videos or add class videos to a public list is also prohibited.  Students registered with the UCO Office of Disability Support Services may request accommodation of the prohibition and must present a copy of the DSS letter to the instructor.

Title IX

The University of Central Oklahoma complies with Section 504 of the Rehabilitation Act of 1973 and the Americans with Disabilities Act of 1990. Students with disabilities who need special accommodations must make their requests by contacting Disability Support Services, at (405) 974-2516.  The DSS Office is located in the Nigh University Center, Room 305. Students should also notify the instructor of special accommodation needs as soon as possible. Per Title IX of the Education Amendments of 1972 (“Title IX”), pregnant and parenting students may request adjustments by contacting the Title IX Coordinator, at (405) 974-3377 or TitleIX@uco.edu. The Title IX Office is located in the Lillard Administration Building, Room 114D.

Important Dates

Week	Dates	Monday	Wednesday	Friday
1	08/17-08/21	Introduction	Hallucinations and Reliable AI	Model Selection and Setting (1) Programming assignment: fundamental python coding with an LLM
2	08/24-08/28	Model Selection and Setting (2)	Prompt Engineering (1)	Discussion and assignment presentation Programming assignment: Model setting & Prompt engineering
3	08/30-09/04	Labor Day	Prompt Engineering (2)	Discussion and assignment presentation Programming: Deploy your own local LLMs
4	09/07-09/11	Introduction to RAG	Data Preparation and indexing for RAG	Discussion and assignment presentation Programming: Building a RAG System
5	09/14-09/28	Building a RAG System	Embeddings & Vector Search (1)	Progress check, discussion, and problem solving
6	09/21-09/25	Embeddings & Vector Search (2)	Exploration of Open-Source Models	Discussion and assignment presentation Programming: Coding with an Open-Source Model
7	09/28-10/02	Deploy a local model and equip it with APIs	Introduction to AI Agents	Discussion and assignment presentation Programming: Equip a local model with OpenAI-like API
8	10/05-10/09	Agentic RAG	Tool Integration with MCP Assignment: Library research	Discussion and assignment presentation Programming: Tool integration with MCP
9	10/12-10/16	Introduction to Skills (1)	Review & Homework Discussion	Fall Break
10	10/19-10/23	Introduction to Skills (2)	Midterm	Discussion and assignment presentation Programming: Equip your system with Skills
11	10/26-10/30	Multi-Agent Systems (1)	Multi-Agent Systems (2)	Discussion and assignment presentation Programming: Build a multi-agent system
12	11/02-11/06	Evaluation of Agentic Systems	Deploying & Monitoring	Progress check, discussion, and problem solving
13	11/09-11/13	Bias, Privacy and Responsible AI	Ethical discussion	Discussion and assignment presentation Final Project: A wheelchair navigation system
14	11/16-11/20	Project development	Project development	Progress check, discussion, and problem solving
15	11/23-11/27	Project development	Thanksgiving
16	11/30-12/04	Review and Discussion	In-Class Demonstration (1)	In-Class Demonstration (2)
17	12/07-12/11			Final exam

This schedule (including exam dates) is subject to change. You are responsible for attending class and staying aware of announced schedule updates.