Skip to content

CS 5090. Software Engineering with Generative AI

Michigan Technological University · Fall 2026 · Dr. Jie "JW" Wu (jie.jw.wu@mtu.edu) Mon/Wed/Fri 12:00 – 12:50pm · Electrical Energy Resources, Room 0227 · Aug 31 – Dec 11 · no final exam

Scaling law drawn as the Quincy No. 2 shaft house


Announcements

  • Oct 16: Bonus: a related workshop for Project 2, Agentic Software Engineering.
  • Oct 13: Short in-class talks from Oct 19 to Nov 13, one or two per class (they replace the Oct 26 topic pitch).
    • What: one or two related papers, then your project pitch (question, motivation, evaluation plan).
    • Length: at least 10 minutes, plus 1–2 minutes of Q&A.
    • Sign up: sign-up sheet. Put your name in one empty row by Fri, Oct 16, 11:59pm. Pairs take both slots of the same day.
  • Oct 5: Midterm 1 moves to Fri, Oct 9 (in class); the Phase 2 debrief is now Wed, Oct 7. The Midterm 1 review sheet and the Week 5 worksheet answers are on the Worksheets page.
  • Sep 14: Project 2 is released: full spec on the Project 2 page. Candidate topics follow on Fri, Sep 18.
  • Sep 4: Project 1 is released: full spec on the Project 1 page.
  • Sep 4: Week 1 slides (Language models & Transformers) are posted on the slides page.
  • Welcome to CS 5090!

Key dates

Date Event
Mon, Aug 31 First day of classes
Fri, Sep 18 Project 2 topics circulated: the research thread starts
Fri, Oct 2 Project 2 research abstract due: context, the gap, your proposed solution
Fri, Oct 9 Midterm 1 (covers Module 1 foundations)
Oct 19 – Nov 13 Project 2 related-work + pitch talks: one or two per class, at least 10 minutes each plus 1–2 minutes of Q&A: one or two related papers, then your project pitch. Sign up by Fri, Oct 16
Fri, Oct 30 Project 1 due in full (one deadline)
Fri, Nov 6 No class (instructor travel)
Sun, Nov 8 Project 2 proposals due: topic pivots allowed until Fri, Nov 20
Wed, Nov 18 Midterm 2 (covers Module 2)
Fri, Dec 4 Project 2 full paper drafts due
Mon & Wed, Dec 7–9 Project 2 design presentations + preliminary results
Fri, Dec 11 Last day of classes; Project 2 final paper due
Dec 14–18 Finals week, no final exam in this course

How the course runs

Two teaching modules with the ⚙ AI-assisted/agentic-SE practice thread woven through both, then a flexible end-of-term block:

  • Module 1 (Weeks 1–8). Foundations. Build a code language model: transformers, training, preferences, reasoning, RAG and agents, evaluation. Runs alongside Project 1 (build → use → evaluate & break → improve your own code LM).
  • Module 2 (Weeks 9–11): ML in Production. Requirements, architecture, quality assurance, operations, responsible engineering: for systems with LLM components.
  • Weeks 12–14. Midterm 2, advanced ⚙ topics from the current literature, then the Project 2 finish: draft → design presentations with preliminary results → final conference-format paper.

The full week-by-week schedule, grading, and policies are in the syllabus. Slide decks are posted on the slides page after each class.


  • Syllabus: schedule, grading, policies (PDF)
  • Slides: posted after each class
  • Project 1: build and evaluate a small code language model (25%)
  • Project 2: research study + conference-style paper (45%)