CS 890 Homepage

CS 890 Schedule

Mon, 01 Jan 0001 00:00:00 +0000

Schedule

Week 1: August 24

Lectures:

Tuesday - Course Introduction + How LLMs Actually Work
Thursday - From LLM to Agent: The Loop, System Prompts, and Tool Calls

Out of Class Activities

Assigned Thursday (due 11:55pm, Wednesday Sept 2): Exercise: Transcript Critique

Week 2: August 31

Lectures:

Tuesday - Claude Code Hands-On: Permissions, CLAUDE.md, Plan Mode
Thursday - Prompting + Spec-Driven Development

Out of Class Activities

Assigned Tuesday (due 11:55pm, Monday Sept 7): Exercise 2: Codebase Comprehension with Claude Code
Assigned Thursday (kickoff due 11:55pm, Sunday Sept 13): Project 0: Personal Knowledge Base

Week 3: September 7

Lectures:

Resources

Mon, 01 Jan 0001 00:00:00 +0000

YouTube Channels / Podcasts

The following channels have consistently good and timely content related to agentic software engineering. I recommend subscribing to them and spending time each week watching their videos. You might also consider looking through their past videos for highly-watched ones or ones on topics that you find interesting.

Matt Pocock

Matt has produced a number of useful things such as his skill library (including the “Grill Me” skilled used in this class), agent terminology, etc. He also has his own for-pay training material that is quite popular. Watch for practical tips and resources.

CS 890 Syllabus

Mon, 01 Jan 0001 00:00:00 +0000

See official syllabus on KSU Course Dog website

Mon, 01 Jan 0001 00:00:00 +0000

John Hatcliff – Bio Sketch

Bio Sketch - To be Completed

Mon, 01 Jan 0001 00:00:00 +0000

Lectures

Part 1: Foundations of Coding Agents

Lecture: Course Introduction + How LLMs Actually Work

This lecture introduces agentic software engineering and explains the LLM machinery beneath coding agents: tokenization, next-token prediction, transformer basics, model training, context windows, and statelessness. Understanding these foundations clarifies how agents can produce code, but also why agent output can be nondeterministic or confidently wrong—and why deliberate context management and verification are essential throughout the course.

Mon, 01 Jan 0001 00:00:00 +0000

Overview

This course introduces students to agentic software engineering: building software and systems with AI agents through disciplined, repeatable, and scalable practices rather than “vibe coding.”

The course begins by teaching students to implement a simple coding agent. LLM-based agents are themselves a computing platform: they operate through a set of primitives that includes calling an LLM with context, invoking tools, and receiving instructions from a human developer. Each step incurs costs and may introduce errors or security risks. To use agents effectively, students need to understand their anatomy, develop a mental model of their execution, estimate costs, and identify potential sources of failure and security vulnerabilities. This understanding also enables students to customize and extend agents for greater effectiveness and efficiency in a particular development context. The course also examines how agents interact with conventional development tools, such as testing frameworks and source code repositories, and how agentic development changes established practices or calls for new tools.

Mon, 01 Jan 0001 00:00:00 +0000

Project

The course contains two individual projecs, and one team project.

Personal Knowledge Base (PKB) (individual project): Use a coding agent to design, implement, and maintain a personal knowledge base with entries in markdown format following Google’s OKF standard. You will use and enhance your PKB throughout the semester to store and take notes on articles, tools, blog posts, YouTube videos, etc. that you find interesting. You will use a markdown viewer such as Obsidian to view and author your PKB content. This project will have an “own it” flavor: once you have the basic functionality set up to meet project requirements, you can design your own enhancements, visualizations in Obsidian, automated workflows (e.g., to have agents grab, summarize, and categorize articles), etc. to make the application fit your needs and whims. Note that this same infrastructure can be used to organize study notes, etc. from other classes and for your research.

Mon, 01 Jan 0001 00:00:00 +0000

Tools

This course will use the following tools, which students should have installed on a personal machine (there will be no provision or lab-based computers). Ideally, you will install on laptops that can be brought to class for in-class exercises.

Coding Agents

Claude Code

You are required to purchase a Purchase Claude Pro subscription (or higher) and have Claude Code installed on a personal machine. The Claude Pro plan is currently $20/month. Don’t ask about substituting any cheaper agents/models, etc. – the answer is “no”. You are certainly welcome to try other things on your own.