Constructor University · AI in CS Education Workshop · June 29, 2026

The New Computing

Education in the AI Era

Hope, Fear, Uncertainty

Computing’s seventy-year dialogue between hope and fear.

Assessments

When AI can complete the assignment, the assignment has to become something only the student can complete.

Courses

New courses that incorporate AI as co-instructor.

Degrees

A new degree for the students today’s programs don’t serve.

Follow Along

geoffreychallen.com/talks/2026-06-29

About the Speaker

Geoffrey Challen

geoffreychallen.com

I love to teach, and I love to build. I teach students to build—with code, and now with AI.

A.B. (Physics) and Ph.D. (Computer Science) from Harvard, supervised by Matt Welsh. Deployed wireless sensor networks on active volcanos. Researched mobile systems at the University at Buffalo, taught operating systems and internet basics, and built the world’s first smartphone platform testbed. Now Teaching Professor at Illinois, teaching computing to thousands of students per year using novel technology and pedagogy.

Resources for Educators

geoffreychallen.com: more about me and my work
learncs.online: free interactive CS1 textbook
cs124.org/educators: course pedagogy and resources for educators
usingandunderstanding.ai/educators: new course on generative AI
computingeducators.org: online community for computing educators

Twenty-Five Years of Code

Four days out of five, for twenty-five years:
writing code by hand to realize my educational visions.

Since 2026 I no longer read, write, or debug code by hand—
and I’m building more than ever.

geoffreychallen.com/essays/healing-agents

Building the New Computing

David opened today with a word: crisis—danger and opportunity in one character. (Apocryphal—but apt.) This is the opportunity.

Assessments

Familiarity with detailed computer coding
is not necessary. Grace Hopper · UNIVAC FLOW-MATIC manual · 1958

In collaboration with

The Traditional Programming Assignment

AI Does the MP

Watch full video (64 min)—sped up 32×

The Programming Assignment, with AI

With or Without AI

With or Without AI?—essay, April 2026

Fall 2025: Blurring the Specification

Homework 3: Linked List
Implement a class LinkedList with these methods:
insert(value)—Add a new node at the end of the list.
delete(value)—Remove the first node with this value. Return true if found, false otherwise.
reverse()—Reverse the list in place.
toString()—Return a string like "1 → 2 → 3 → null"
Handle edge cases: empty lists, single-element lists.

Homework 3: 🔗 List
Build a 🔗 with these powers:
➕—Stick a new 📦 at the 🔚.
❌—Find & destroy the first 📦 matching this. ✅ if found, ⛔ if not.
🔄—Flip the whole thing 🙃.
🖨️—Spit out "📦 → 📦 → 📦 → ∅"
⚠️ Watch out for: 📭 and lonely 📦.

Homework 3: 🔗 List
Build a 🔗 with these powers:
➕—Stick a new 📦 at the 🔚.
❌—Find & destroy the first 📦 matching this. ✅ if found, ⛔ if not.
🔄—Flip the whole thing 🙃.
🖨️—Spit out "📦 → 📦 → 📦 → ∅"
⚠️ Watch out for: 📭 and lonely 📦.

Fall 2025: A Failed Experiment

“The project is too difficult to do by yourself, yet too easy to do with Claude.”

“My flow for working on the assignment ended up being: 1. Hey Claude, look at X test and expand it to make a comprehensive test suite. 2. Hey Claude, now write the code to pass the tests!”

“I believe we all had Claude do all of the coding for each assignment.”

“You can just ask Claude to do everything for you.”

Spring 2026: An Independent Project

cs124.org/ai

Time on Task Grading

What They Built

Best Assistive Impact

Simplified Senior Launcher

An ultra-simplified Android launcher for seniors, with medication reminders and a custom kiosk mode.

“Separates a student project from a life-changing tool.”

Best Research Workflow

Lab Experiment Assistant

Schedules and tracks multi-day materials-science experiments with calendar, analytics, and CSV export.

“Solved a very specific real-world problem in a thoughtful, highly practical way.”

Best Immersive Experience

Emotion-Based Quote App

Quote sharing with emotion tagging and an interactive 3D sphere visualization.

“Genuinely creative idea unlike anything else in this class.”

cs124.org/duckies

Teaching Classical Programming

learncs.online/solve

Frequent Small Assessment

High Stakes (4 hours proctored)

Final

Frequent Small Assessment (15 hours proctored)

Q10

Q11

Q12

Q13

Q14

Retake Windows

W10

W11

W12

W13

Dropping Low Scores

Institutional Support

Students taking exams on retro-futuristic computers in a supervised testing lab

cbtf.illinois.edu

Courses

Should the computer program the kid,
or should the kid program the computer? Seymour Papert · in Alan Kay, A Personal Computer for Children of All Ages · 1972

In collaboration with

Built Through Conversational Programming

usingandunderstanding.ai/create

Conversational Assessment

Oral Assessment at Scale

AI interviewer

↔conversation

Student

An oral exam: you can’t fake understanding in a real conversation.

Scalable, high-frequency, auditable.

What They Built

Tank Hub

Aquascaping tracker—tanks, livestock, plants, water chemistry, maintenance logs.

Phi Sigma Sigma

Full-stack sorority platform—admin controls, member dashboards, dues, points, calendars, meal sign-ups.

My Hypothesis

Stacked

Classical programming

↓

AI collaboration

Independent, but related

Classical programming

↔

AI collaboration

AI collaboration is a distinct skill that is independent from—but related to—classical programming.

geoffreychallen.com/essays/another-skill

“Conversational Programmers”—Not a New Idea

Then

Learn enough programming to talk to coders: literacy to communicate, not to produce.

Chilana (2015–16) · Guzdial · Cunningham · Hur & Cunningham, ICER 2024: the single most popular endpoint for non-CS majors who take a computing course.

Now

Direct coding agents in plain language, and own the intent, the spec, and the verification.

They learned to talk to the people who wrote the code. Now the code writes itself, and the loop of asking, checking, and iterating is the whole job.

Same instinct: be precise about software you don’t hand-write. The partner changed from a person to a machine.

Closing Software’s Oldest Split

Cares about the problem

domain expert · knows what’s right

lossy
requirements
channel

Can change the artifact

developer · holds the keys

For decades the person who cared and the person who could build were two different people. Most of our security and usability failures live in that gap—not negligence, a broken channel.

One person, directing an agent

The person who cares can finally change the thing. Evaluation was always the domain expert’s job—it was just stranded on the wrong side of the wall.

The Trust Gradient

      How to make it true: the agent’s
      yours: what should be true here
    

Trust, verify

Syntax, scaffolding, secrets, input sanitization. The agent is boringly reliable.

The traps

Green checks that lie. 100% coverage of the code it wrote; WCAG AA a real user still can’t navigate.

Yours to own

Threat model, user model, domain invariants, what’s worth building.

The skill is epistemic discipline—knowing where to trust, where to verify, and what only you can own. We grade the delta: what the student adds past what the agent raised.

“Epistemic discipline” borrowed from Paolo Ciancarini, The conversation is the computer (this workshop).

A Studio, Not a Lecture Hall

Borrowed from the architecture studio: one artifact, many crits—each pass foregrounds a different concern, dragging accountability one zone rightward along the gradient.

Build

demo · delete

Build

demo · delete

Build

demo · delete

→

One final project

user-anchored · carried across crits

A portfolio of disposable projects. Build, demo, delete: throwaway work can’t over-design. The portfolio is also a search: you keep the one with legs.

The return is subtractive. A real user model usually says you don’t need 70% of what you built. The end is where the project gets right, which means smaller.

Fall 2026: Conversational Programming

If you can talk it, you can create it. Conversational Programming Manifesto

Degrees

By augmenting human intellect we mean increasing the capability of a person
to approach a complex problem situation, to gain comprehension, and to derive solutions. Douglas Engelbart · Augmenting Human Intellect · 1962

In collaboration with

Two Computing Populations

“Fascinated. They want to know everything about everything for the sake of knowing.”

Faculty participant · Future of Computing

“Driven. They had a vision—and were interested in everything they thought would help with that, and nothing that wouldn’t.”

Faculty participant · Future of Computing

Computer Science

Computing as ends

Applied Computing

Computing as means

Applied Computing

Explore · Design · Prototype · Critique · Iterate · Validate.

Current proposal—working draft

The New Core

Y1 Fall

Conversational
Programming

Computing
in Culture

Y1 Spring

Agentic Software
Development

Y2 Fall

AI Models
and Agents

Integrative Design
Studio I

Y2 Spring

How Software
Works

Conversational Programming. Designing and building software through conversation with AI—a studio in spirit before the formal studios begin.

Computing in Culture. How software has reshaped human behavior. The critical lens runs in parallel with learning to build.

Agentic Software Development. Constructing and evaluating agents, agentic workflows, real deployment. Students learn to be effective directors of AI.

AI Models and Agents. How models actually work. Placed after fluency: students step back to reason about systems they’ve already built with.

Integrative Design Studio I. Students bring a paired domain problem into the studio and design software around it. Architecture-style critique.

How Software Works. The classical programming course—what happens beneath the conversational layer. Doubles as a bridge into CS upper-division electives.

Current proposal—working draft

Design Pillars

Domain Concentration

Any existing minor on campus. Each student braids in another field—biology, business, design, the humanities. Ideas and problem-understanding drive the work; the domain tells you what to build.

Formation

Writing, literature, moral reasoning, studio art. Professional capabilities, not decoration. Students who build for people need to communicate, reason ethically, and see clearly.

Studio Progression

A portfolio model: Integrative Design Studio I → II → III, then a capstone with public defense. Students graduate with a body of work, not a transcript—scope and defense grow with each studio.

Borrowed from architectural pedagogy. Tested for centuries.

We are as gods, and might as well get good at it. Stewart Brand · 1968

We are as gods, and have to get good at it. Stewart Brand · 2009

History Rhymes

“I decided data processors ought to be able to write their programs in English, and the computers would translate them into machine code.”
Grace Hopper
FLOW-MATIC, c. 1955

“Working within the loose constraints of predetermined strategies, computers will in due course be able to devise and simplify their own procedures for achieving stated goals.”
J.C.R. Licklider
Man-Computer Symbiosis, 1960

“We at Dartmouth envisaged the possibility of millions of people writing their own computer programs.”
John Kemeny
Man and the Computer, 1972

“It is practically impossible to teach good programming to students that have had a prior exposure to BASIC: as potential programmers they are mentally mutilated beyond hope of regeneration.”
Edsger Dijkstra
EWD498, 1975

“Since we do not now have any ways of making computers wise, we ought not now to give computers tasks that demand wisdom.”
Joseph Weizenbaum
Computer Power and Human Reason, 1976

1950

1960

1970

1980

1990

2000

2010

2020

1958

1960

1972

1975

1976

2022

2025

Closing Thoughts

1
Change our work, and ourselves. New assessments, new courses, new degrees, and educators who keep learning.
2
Use the tools to know the tools. You only learn what they can do by building with them.
3
Embrace the uncertainty. AI collaboration is a new skill, and we don’t yet know how to teach it. We need a lot of people trying a lot of new things.
4
Deliver the future. Tech without the tech companies: the ultimate fulfillment of the dreams of computing’s pioneers.

Thank You

Excited to keep building this together. Happy to take questions now, and to keep the conversation going by email after.

Join our online community for computing educators: computingeducators.org

Thank you to David Rosenblum for the invitation, Giancarlo Succi and the School of CS & Engineering, co-chair Jürgen Schönwälder and organizer Stella Byun, and everyone in Bremen and online.