Skip to content
Foundations for bootcamps · teach coding in the AI era

The Coding Bootcamp Curriculum For The AI Era.

Your students will code alongside AI. Foundations has them predict, write and explain code first, and AI unlocks only after they've tried. You share one join code and see each student's lessons and tasks completed before the next session.

Inside Arclab right now
Built for the AI eraStaged AI, enforced on the serverPython, Java and Next.jsOne join code per cohortBrowser onlyNo upfront cost
How It Works

Create A Cohort. Share A Code. Teach With Data.

Arclab Foundations learner dashboard: the Cohort tab with a field to enter the join code an organizer shared
Arclab Foundations learner dashboard: Python Foundations, Java Foundations and Junior Developer: Next.js track cards
Arclab Foundations: Compare with AI blocked with the message Complete the earlier lessons before using AI here
Arclab Foundations: the Predict task marked Correct and the next Modify task waiting in the lesson panel
1.

Create a cohort

Name it and get a six-character join code. We set up your organizer account with you.

2.

Students join and start

They sign in, enter the code under Cohort and pick a track. Nothing to install, any modern browser.

3.

They learn, AI waits

Every lesson is predict, write, run, explain. AI help opens in stages and only after the earlier lessons are done.

4.

Teach with the data

See each student's lessons and tasks completed. When a cohort finishes, move them into the Sim for a first week on a team.

What Students See

The Workspace Your Students Get.

A dark editor that works like VS Code, the lesson beside it, a terminal underneath, and an AI mentor that waits its turn.

Arclab Foundations: a learner's area function in main.py and the in-browser terminal printing The room is 20 square meters
Real code, running in the browser. Students write main.py and run it in a terminal. Python runs in the browser, so there is nothing to install or maintain.
Arclab Foundations: the AI mentor panel open beside the editor, saying it won't give answers directly
A mentor that teaches, not tells. It explains concepts and errors and asks questions back. It doesn't hand over answers.
Arclab Foundations: Compare with AI blocked with the message Complete the earlier lessons before using AI here
AI that respects the syllabus. Ask for AI too early and the server says no: complete the earlier lessons first.
Arclab Foundations Next.js track: page.tsx in the editor and a live preview of the app running in the browser
Next.js with a live preview. For web cohorts: a real Next.js project boots in the browser, with the preview beside the code.
Arclab Foundations Java track: Hello, machine lesson with Predict and Modify tasks marked correct
Java for CS-style programs. The same sixteen modules in Java, for programs that teach typed languages first.

Tell Us About Your Next Cohort.

Tracks, size and start date. We'll have you set up before it begins.

Talk to us →
Why Now

Your Students Already Have AI.

Any student can paste a task into a chatbot and hand in working code. That doesn't tell you whether they understand it. Foundations is built around that: students predict and write first, explain back in their own words, and only then get AI, in stages you don't have to police.

chatbot · bill_splitter.py · written in 2s
def split(total, people):
  return round(total / people, 2)
✓ donepasses. do they get it?
  • predict it first
  • write it yourself
  • explain it back
  • find the bug
  • then compare with AI
  • review what AI wrote
AI Policy

An AI Policy You Don't Have To Police.

Every module carries an AI phase, and the server enforces it. In the Python track that means seven modules of no AI, then explain, then compare, then collaborate. Your students build the basics themselves before AI joins in.

  1. 1No AIPython modules 1 to 7

    You write every line yourself. Hints are there when you're stuck, answers are not.

  2. 2AI explainsmodules 8 and 9

    Ask the mentor what a concept or an error means. It explains; it won't write the code for you.

  3. 3AI comparesmodules 10 to 13

    Once your code passes, open an AI version beside yours and work out what it did better, worse or wrong.

  4. 4AI collaboratesmodules 14 to 16

    Build with AI the way the job works now, then read and review everything it wrote.

Java opens AI at module 14 and Next.js at module 8. Want a different schedule for your cohort? Tell us.

For Instructors

What You Get.

Simple terms ›
cohort · join
Have a join code?

 A7K2QX

[ Join cohort ]
one code per cohort

Cohorts and join codes. One six-character code per cohort. Students enter it once and they're in.

organizer view
cohort · students
student     lessons  tasks
Student A        12     41
Student B         9     30
Student C         2      5
check in with C

Progress per student. Lessons and tasks completed for each student, so you know who to check on before the next session.

organizer view
syllabus
python   16 modules  215 tasks
java     16 modules  201 tasks
next.js  23 modules  190 tasks
ready to teach

Curriculum you don't write. Python, Java and Next.js: 215, 201 and 190 tasks, with five kinds of task in every lesson.

3 tracks · 55 modules
ai policy · python
modules 1-7    no AI
modules 8-9    explains
modules 10-13  compares
modules 14-16  collaborates
checked on the server

Staged AI. AI unlocks by module and the server checks it. No honor system.

no AI → explain → compare → collaborate
mail · hello@arclab.app
to: hello@arclab.app
re: our spring cohort

Could we add a module on pandas?
we build it with you

Custom modules. Need extra exercises, a different order or a module for your stack? Email us and we build it with you.

on request
sim.arclab.app · #team
Mon 10:58
Maya: welcome to Seatly!
Maya: your first ticket is up
SEAT-12 party too big for table
week one on a team

A capstone after. Graduates can go straight into the Sim, a junior engineer's first week on a real codebase.

sim.arclab.app
Try A Task

One Task, As A Student Sees It.

foundations / 08 Collections / main.pytask 1 of 3 · AI locked

Predict the output before you run it.

nums = [3, 1, 2]
print(sorted(nums)[-1])
The Syllabus

The Syllabus You Can Teach Tomorrow.

Three tracks, each a straight line from the first lesson to a capstone, with the AI phase of every module. Build modules are the projects students make along the way.

Python Foundationsbeginner · 16 modules · 72 lessons · 215 tasks
  1. 01Make the machine do somethingno AI
  2. 02Data & variablesno AI
  3. 03Build: Vibe Check Quizno AI
  4. 04Making decisionsno AI
  5. 05Build: Smart Bill Splitterno AI
  6. 06Repetitionno AI
  7. 07Build: Your Year Wrappedno AI
  8. 08CollectionsAI explains
  9. 09Build: Expense TrackerAI explains
  10. 10FunctionsAI compares
  11. 11Build: Brainrot TrackerAI compares
  12. 12Thinking in objectsAI compares
  13. 13When things go wrongAI compares
  14. 14Coding with AIAI collaborates
  15. 15Reading other people's codeAI collaborates
  16. 16CapstoneAI collaborates
Java Foundationsbeginner · 16 modules · 72 lessons · 201 tasks
  1. 01Make the machine do somethingno AI
  2. 02Data & variablesno AI
  3. 03Build: Vibe Checkno AI
  4. 04Making decisionsno AI
  5. 05Build: Bill Splitterno AI
  6. 06Loopsno AI
  7. 07Build: Year Wrappedno AI
  8. 08Collectionsno AI
  9. 09Build: Expense Trackerno AI
  10. 10Methodsno AI
  11. 11Build: Brainrot Trackerno AI
  12. 12Thinking in objectsno AI
  13. 13When things go wrongno AI
  14. 14Coding with AIAI collaborates
  15. 15Reading codeAI collaborates
  16. 16CapstoneAI collaborates
Junior Developer: Next.jsafter Python · 23 modules · 71 lessons · 190 tasks
  1. 01From Python to the webno AI
  2. 02JavaScript fundamentalsno AI
  3. 03Your dev toolkitno AI
  4. 04HTML & JSXno AI
  5. 05Styling with Tailwindno AI
  6. 06Build: Profile Cardno AI
  7. 07TypeScript basicsno AI
  8. 08React components & propsAI explains
  9. 09State basicsAI explains
  10. 10Pages & routingAI explains
  11. 11Build: Personal PortfolioAI explains
  12. 12Advanced state patternsAI explains
  13. 13Build: Interactive QuizAI explains
  14. 14Async JavaScriptAI explains
  15. 15The server sideAI compares
  16. 16Connecting to SupabaseAI compares
  17. 17Build: Live DashboardAI compares
  18. 18Forms & full CRUDAI compares
  19. 19Build: Bookmark ManagerAI compares
  20. 20Authentication & protectionAI compares
  21. 21Build: Team PlannerAI compares
  22. 22Ship it & AI masteryAI collaborates
  23. 23Capstone: your app, your rulesAI collaborates
How It Compares

Why Not Slides And A Chatbot?

FeatureArclab FoundationsSlides + a quiz toolStudents + a chatbot
Students run real code in the browser-
Checks understanding, not just outputquizzes-
AI unlocks in stages, enforced on the server--
One join code puts students in your cohortsometimes-
Lessons and tasks completed per studentquiz scores-
Python, Java and Next.js ready to teachyou write it-
Upfront costNoneVariesSeats

The other columns describe common setups, not any one product.

Simple Terms.

Upfront costNone
PricingPer active learner, per month
Nonprofits and community programsFree
ContractMonth to month
Student requirementsA modern browser
SupportEmail and a shared channel

Questions.

What does it cost?

No upfront cost; per active learner per month. Nonprofit and community programs are free.

What do students need?

A modern browser and a Google account. Nothing to install.

How do students join my cohort?

You share the cohort's six-character join code. Students sign in, open Cohort in their dashboard and enter it.

What can organizers see?

Each student's lessons completed and tasks completed, per cohort. Not the content of mentor conversations.

Can students just ask AI for the answers?

Not inside Foundations. AI is off for the first modules and opens in stages, checked on the server, and only after earlier lessons are done. Tasks also ask students to predict and explain, which a pasted answer doesn't cover.

Which tracks can we teach?

Python and Java for beginners, and Next.js for web cohorts who know some Python. Each runs in the browser.

Can we use our own curriculum?

Email us and we'll build it with you: extra exercises, a different order, or a module for your stack.

Which cohort sizes work best?

Six to a few hundred. The view is built for one instructor following a class of 20 to 40.

Cohorts That Finish.