Learning to Code When AI Does the Heavy Lifting

workshop-cursor-deck.ai-builders.space

Yuzheng Sun · April 1, 2026 · Room Willow A

Founder · Meta / Amazon / Statsig / Tencent

Build with AI: Mindset, Habits & Tools for the Next 20 Years

⭐ 4.9 (72 reviews) · Next cohort at the time: Apr 13–27, 2026

maven.com/superlinear/aibuilders →

PART 1 How to Use Cursor

◉ DEMO Live Cursor Demo — open your machine

PART 2 How to Debug

PART 3 How to Write Effective Prompts — the 3C Framework

WRAP-UP Your Next Step + Bonus

Sound familiar?

"I use ChatGPT every day but it doesn't feel like a real breakthrough"

"AI gives me stuff I still have to rewrite — might as well just do it myself"

"I've heard about Cursor, but I don't know where to start"

30% most people

→ 10×

a few

The difference isn't the tool. It's the method.

"In my lifetime, I have seen two demonstrations of technology that struck me as revolutionary."

— Bill Gates, March 2023, after seeing GPT-4

The First: GUI

The graphical interface put computers in the hands of a billion people. PCs, software, the internet, smartphones — all built on this paradigm shift.

The Second: AI

"The age of AI has begun." Not a new tool — a new era. Same magnitude as the GUI.

Why this matters:

• Before GUI: only experts used computers
• After GUI: anyone could
• With AI: anyone can now harness compute
• What changes: what ordinary people can do

When the electric motor was invented, factories simply replaced the steam engine with it — everything else stayed the same. Efficiency improved by 30%.

Decades later, someone realized: motors can be small. You don't need to design a factory around a central power source — you can design it around the production process itself.

The assembly line was born. Efficiency jumped tenfold.

Today's AI is the same story.

What most people do: keep the same workflow, just "ask AI" at one step. — That's putting the electric motor where the steam engine was.

The real transformation: redesign how you work around what AI can actually do.

ChatGPT is your advisor. Cursor is your engineering team.

Dimension | ChatGPT (chat mode) | Cursor (agentic mode)

Context | Every conversation starts from zero — you re-explain everything each time | Reads your entire project: files, rules, history — understands your setup

Persistence | Output lives in the chat window — disappears when you close it | Output goes into actual files — it accumulates, it's reusable next time

Self-correction | Only updates when you say it's wrong — silent on errors you don't notice | Has a real execution environment — runs code, sees errors, fixes itself

This isn't a feature comparison. It's two fundamentally different ways of working — one where you drive AI; one where AI drives, and you review outcomes.

Traditional: Process Certainty

• Prompt → check → edit manually → repeat
• AI output capped by your review bandwidth
• You're the bottleneck

AI-Native: Outcome Certainty

• Define "what good looks like"
• Give AI rules + context
• AI runs, checks, iterates → you review
• Highest leverage: define the finish line

When AI can produce far more than you can review line-by-line, the bottleneck isn't attention — it's how clearly you defined "done."

You don't need to Already know how to code Understand every AI concept Learn everything today You just need One real task to work on To tell Cursor your goal and context To watch it run — then review

Red errors are clues, not disasters.

STEP 1

Read the error — it's a clue

• Not a judgment — it's a message
• Tells you what broke and which line
• Read it before asking Cursor anything

STEP 2

Give Cursor the full picture

• Don't just paste the error
• "I expected X. Got this instead."
• Expected + actual = the right fix

STEP 3

Let Cursor run, check, and fix

• Let it execute, see output, and iterate
• Ask: "Why did this break?"
• Not just: "Does it work now?"

The goal isn't to never break things. It's to recover confidently and learn faster when you do.

A vague instruction can produce code with no errors — and still lose your data.

Download link:

ai-builders.com/resources/lesson4_data.zip

⬇ Download lesson4_data.zip

We'll use this dataset to practice turning vague instructions into precise, AI-ready specs with the 3C Framework.

You have a photo folder. Two files named sunset.jpg — one ocean, one mountain. You ask Cursor:

"Write a script to move all photos into one new folder, flatten the subdirectory structure."

What AI assumes

Same filename = same file. Overwriting is fine. You didn't say otherwise.

What actually happens

One sunset.jpg is silently deleted. No error. No warning. No trace.

THE TRAP

The code ran without errors. It did exactly what you asked. But you lost data. Vague requirements produce technically correct code that's a wrong product.

The problem isn't that AI is dumb. It's that you carry implicit assumptions that AI doesn't share.

About identity — you assumed:

You: "Photos with different content are different files." AI: "Photos with the same filename are the same file."

About conflict — you assumed:

You: "Overwriting would lose data — obviously bad." AI: "If same filename = same file, then overwriting loses nothing."

The skill is not writing — it's excavating.

• Surface hidden assumptions
• Turn implicit decisions → explicit rules
• That's what Criteria is for

C1 · CONTEXT

What's the #1 priority?

• Set the north star first
• Data safety: better to do nothing than lose a file
• This overrides all other decisions

C2 · COMPONENTS

What are the steps?

• Scan source directory (recursive)
• Identify duplicate files
• Execute file operations
• Generate verification report

C3 · CRITERIA

Make implicit rules explicit

• Duplicate? → content hash, not filename
• Name conflict? → auto-rename, never overwrite
• Safety? → copy-only, originals stay
• Which files? → .jpg / .png only

Data Consistency Check

• Add to end of every script
• input = output + duplicates + skipped
• If unequal → print RED WARNING

Visual Acceptance Report

• Generate report.html
• Left: files kept · Right: duplicates removed
• One human scan > 100 automated checks

THE MINDSET SHIFT

• "No errors" ≠ correct output
• Look for data integrity, not just no red text
• Define "done" before you run

LEVEL 1

Only used ChatGPT

• Open Cursor on one weekly task
• Let it read your actual files
• Notice the context difference No coding required. Just feel it.

LEVEL 2

Cursor but prompt one-by-one

• Write your standards in a file
• AI reads rules before acting
• Output quality jumps immediately

LEVEL 3

Already building workflows

• Add automated acceptance checks
• Second agent reviews / script validates
• "It works" → "It's repeatable"

Shared starting point at every level: write down what "good" looks like before you ask AI to do anything.

The process of making this deck practiced every principle we just covered.

Step | Human did | AI did

1 · Brief | 1 message: workshop info + agenda + lesson content to include | Read original 747-line deck + lesson 4 materials + workshop screenshot

2 · Content | Zero lines of HTML written | Output 900+ lines of HTML: 20 slides, full design system retained

3 · Adaptation | Specified: English, debug section, 3C framework, Tencent in EN | Restructured narrative arc, added 3 new sections, repurposed all assets

1 message

→ 912

lines of HTML

~10 min

human time

vs 20 slides

complete deck

This is "outcome certainty" in a real workflow: human defines the goal and constraints, AI handles all execution — including the sentence you're reading right now.

DOWNLOAD THE APP

Superlinear Academy

Build useful things with AI

• AI Builder courses & projects
• Community feed + share your work
• Lifetime Q&A with instructors

⬇ Download on App Store

COMMUNITY PROJECTS

What other builders have made:

🗺 EvoMap — collective AI memory for Agents

📊 Career Signaling — AI-powered job tracker

🤖 Personal AI assistant for daily workflows

Browse All Projects →

• Learn AI for 20 years, not the next 20 days
• Stop treating AI like a chatbot — design the context
• Build a meta-learning system that compounds over time
• Real projects, real code, real feedback from instructors

RATING

⭐ 4.9 / 5.0 NEXT COHORT AT THE TIME

Apr 13–27, 2026

REVIEWS

72 students

Enroll at maven.com/superlinear/aibuilders →

github.com/sunyuzheng/workshop-cursor-deck

Fork it · Adapt it · Build your own AI workshop

Every slide, every design decision, every prompt used to build this deck is documented in the README — a live demo of the process.

Context · Components · Criteria. The difference is method, not tools.

Yuzheng Sun · Superlinear Academy · Seattle (AI) Startup Summit · 2026