For program advisors

The full 12-week curriculum, written to be lived in the room.

Session-by-session plans for the advisor running a VIBE cohort, whether you are a teacher, a counselor, or a community volunteer who uses AI in your work. The arc covers what generative AI actually is and how it works, how industries are quietly being reshaped by it, how a student moves from a raw idea through honest due diligence into a real shipped piece of work, and how to think about audience, scale, and cost the way a working professional would. The curriculum is tool-agnostic by design, and it is built to flex around any school’s software and hardware restrictions. Adapt it freely, since this is a framework, not a script.

How to use this

Read it once, then adapt it for your school and community.

Each week is one session, ideally somewhere between 60 and 75 minutes if your school can carve out that block, and 45 minutes at the very minimum if it cannot. Every session has a clear goal, a time map you can follow or adapt, activities you can run as written, and advisor notes that capture the lessons we have learned in real cohorts. The curriculum works with any approved AI tool and with any device your school supports, including Chromebooks, iPads, and laptops.

Tool-agnostic

Activities work with whichever AI platform your district has already approved. We name specific examples to make planning easier, but nothing on the list is mandatory.

Hardware-flexible

Every session includes a plan for schools that work in restricted environments, and chat-only AI is genuinely enough for most of the program.

Yours to adapt

Times, orders, and activities can all shift to match your community. What truly matters is the arc, which moves students from try, to imagine, to build, to ship.

A note on session length. The curriculum below was written for a 60 to 75 minute weekly session, which gives students enough time to genuinely build, reflect, and share. Many schools start with 45 minutes during their first cohort and find that a longer block becomes possible by semester two as the program proves itself. If you only have 45 minutes, run the “tight” version of each week by trimming the share-out and lengthening the build window.
Parents and guardians are welcome in the room. A parent who would like to sit in is genuinely welcome at any session, and parental presence has a noticeable positive effect on cohort culture. Parents become part of the support system, they see what their student is actually doing each week, and they often pick up the same skills alongside the kids. The advisor coordinates which sessions are most parent-friendly to attend.
AI access timing. Free-tier AI tools are sufficient through the early weeks of the program. Once students enter the build phase, a paid subscription to an AI platform that supports real project work, typically around $20 per month, becomes useful for the work students do at home between sessions. This is generally paid by the family. Scholarships through the VIBE Fund exist for families where the subscription would be a hardship.
Between-session work is encouraged. Real familiarity with these tools comes from spending time with them outside the room. In the early weeks, students often leave with a small idea or prompt to think about before the next session. Once the build phase begins, students are encouraged to keep building on their own project at home, at their own pace. Nothing is graded and nothing is forced, but cohorts where students engage between sessions consistently ship more confident work.

Before week 1. Confirm your school’s AI tool list with the technology coordinator, review the AI acceptable use policy, collect parent consent forms, brief any guides you have lined up (the program runs without them too if your community recruitment is still in motion), and decide which all-in-one platform you will introduce in the very first session (Lovable, Replit, or Bolt are excellent kid-friendly choices). See the fall and spring calendar for help picking your start date.

V Venture Weeks 1–3

Explore the landscape.

Students try the tools, iterate, and meet someone who builds with AI for a living. The goal is comfort and curiosity, not mastery.

Week 1

What is generative AI, and your very first build.

Goal: every student leaves the first session with a basic mental model of how generative AI works, and with something they themselves made on a real platform. Comfort and curiosity matter far more than polish.

  • 0–5Welcome
  • 5–15What is generative AI
  • 15–25Platform demo
  • 25–55Hands-on build
  • 55–65Share
What happens
  • Advisor introduces VIBE in two minutes: one project, twelve weeks, one real thing shipped by each student. Frame the program as learning to direct AI, not learning to write code.
  • Plain-language explainer of what generative AI actually is. Use this rough script if it helps: “Generative AI is a kind of software that has read a huge amount of human-made text, images, music, and code, and learned to produce new pieces in the same style when you describe what you want. It is confident, it is fast, and it is wrong often enough that you still need a person to direct it. That is your job in this program.”
  • Advisor does a live demo on the platform the cohort will use for the rest of the program. Lovable and Replit are both excellent kid-friendly all-in-one platforms (Bolt is another solid option), and we recommend picking one and sticking with it. Pick something simple to demo, like a webpage with the student’s name on it or a tiny quiz, then prompt the AI in front of the room and walk through the whole messy process out loud, including the wrong outputs and the corrections.
  • Each student picks one thing to make, whether that is a short story, a simple webpage, a quiz, a business idea, a song lyric, a piece of art, or a small game, and they use the platform to make it real on their own screen.
  • Last ten minutes: each student shares one sentence or a screen of what they made today.
Advisor notes. Show your mistakes in the demo on purpose. Students are watching to see if AI is magical or usable, and your job is to make it look usable. Do not filter ideas at this early stage. “Something on screen” is the bar, and it does not need to be good yet.
Flexibility. The session works with any chat AI on any device, including a chat-only fallback if your school cannot approve Lovable, Replit, or Bolt before the first session. Students can use AI to write, design, plan, draw, or describe, not just to write code.
Week 2

Industries already changing, and the conversation rather than the command.

Goal: students see why this skill matters across industries, and they learn that iterating with AI through real conversation produces dramatically better results than single prompts.

  • 0–5Check-in
  • 5–20Industry view
  • 20–35Prompt demo
  • 35–60Improve
  • 60–65Share
What happens
  • Quick check-in. Ask two or three students what they thought of week 1, and listen for what surprised them.
  • Industry view. Walk students through the real picture of where AI is already showing up in working life. Software, design, marketing, healthcare, law, customer service, education, film, music, and the trades all have working professionals using AI tools every day, and the strengths students should know about include drafting written work, generating images and video, summarizing dense material, writing and debugging code, analyzing data, and acting as a thoughtful sounding board for ideas. The lesson is honest. The tools are already inside many industries, the gap between what the tools can do and what most people are using is real, and learning to direct these tools early is a meaningful head start.
  • Pivot to the language lesson. AI is a collaborator and not a vending machine, and the conversation matters more than any single command.
  • Demo the same request as a bad prompt, a good prompt, and a great prompt side by side. Show how context, constraints, and examples change the output. Then demo something most students have never seen, namely asking the AI a question to draw out more information you did not know you needed, the same way a curious junior employee would ask a senior coworker.
  • Students take their week 1 project and make three specific improvements using better prompting and at least one back-and-forth conversation with the AI.
  • Each student shares one improvement they made and one new question they asked the AI along the way.
Advisor notes. The key lesson is that context is everything, and the more a student tells the AI about what they want, the closer it gets to delivering it. Students who felt unsatisfied with week 1 should leave week 2 feeling noticeably better about their project. The industry view in particular helps students see that this is not just a tech-kid program, since AI is reshaping nearly every kind of work.
Flexibility. The whole session is a pure chat activity that works with any text AI on any device, with no installation needed.
Week 3

First guide visit (or AI in the wild).

Goal: Students see what building with AI looks like in a real professional's life.

  • 0–5Intro
  • 5–20Career + AI
  • 20–35Q&A
  • 35–45Follow-up
Option A, with a guide available
  • Advisor introduces the guide: name, field, one sentence about how they use AI.
  • Guide shares their actual work: not a TED talk, a show-and-tell. What they build, how AI helps, where it fails.
  • Open Q&A, where students ask the guide anything they want.
  • Last ten minutes: students with related interests get brief follow-up time.
Option B, without a guide
  • Advisor leads an “AI in the real world” walkthrough using one short article, podcast clip, or video of a working professional in any field discussing how AI shows up in their job.
  • Students share what surprised them and what they want to dig into more.
  • Each student names one field they are curious about and writes one sentence on how AI might show up there.
Advisor prep. If you have a guide, send the guide kit three days before and have each student prepare two questions beforehand. If you do not, line up one short clip or article from a real practitioner before the session. Either path delivers the same lesson: that AI is a tool people are using in real work right now.
Flexibility. No tech required. Screen-sharing is optional. Guide visits work by phone, video, or in person depending on what your school allows, and the no-guide path needs only a single playable clip or printed article.
I Imagine Weeks 4–6

Find direction.

A second field. A project idea. A plan a student can actually finish. This phase turns exposure into intention.

Week 4

Second field, second perspective.

Goal: Students encounter a completely different field and see how AI applies there.

  • 0–5Intro
  • 5–20Career + AI
  • 20–35Q&A
  • 35–45Follow-up
Option A, with a guide available
  • Same structure as week 3, with one key difference. Pick a guide from a very different industry this time. If week 3 was a software engineer, invite a nurse, a designer, an artist, or a tradesperson instead.
  • The contrast between guides is the lesson. After two very different fields, students start to see that working with AI is a universal skill, not a technical one.
Option B, without a guide
  • Pick a field that contrasts sharply with whatever week 3 covered. Use a short video, podcast clip, or article featuring a working professional discussing AI in their craft.
  • Students compare how AI shows up across the two fields and discuss what is similar and what is different.
  • Each student names one industry they want to know more about, and the advisor offers a starting clip or article they can explore on their own time.
Site visit option. If a guide invites the cohort to their workplace, this is worth doing. The school's existing field trip policies apply, including a school staff member attending the trip per those policies.
Flexibility. If in-person visits are restricted, video or phone call formats still work. A 20-minute call with a real practitioner in a different field is powerful, and the no-guide path needs only a single playable clip or printed article.
Week 5

From a ball of clay to a vetted idea, with role simulation and a focus group.

Goal: every student leaves with at least one real project idea that has been pressure-tested by the cohort, by the AI in different roles, and by their own honest reflection on whether anyone actually wants this thing.

  • 0–5Frame
  • 5–20Brainstorm
  • 20–35Napkin pitch
  • 35–55AI focus group
  • 55–65Reshape
What happens
  • Advisor frames the challenge: “You have six weeks to build something real. What would you want to exist in the world that does not yet, or what would you want to make better?” Frame the work as molding a ball of clay into a real, mature idea, and remind students that a great idea almost never starts great.
  • Fifteen-minute free brainstorm. Students write every idea without filtering anything out, and nothing is too big or too small for the page yet.
  • Napkin pitch. Each student writes three sentences about their best idea, covering what it is, who it is for, and what problem it actually solves for that person.
  • AI-assisted due diligence and focus group. Each student opens a chat with the AI and runs three short conversations to vet their idea. First, ask the AI to play the part of three different potential users (for example, a 14-year-old classmate, a busy parent, and a small-business owner) and react honestly to the pitch. Second, ask the AI to play the part of a skeptical investor or critic and poke holes in the idea. Third, ask the AI to suggest two ways the idea could be reshaped into something genuinely unique or more valuable. Capture every useful response in a notes doc.
  • Reshape. Students take everything the focus group surfaced and write a second, sharper napkin pitch that names what changed and why.
Three due-diligence questions every student should answer
  • Is this idea genuinely unique, or am I copying something that already exists? If it exists, how is mine different in a way that actually matters to a real person?
  • Where is the value? Who is going to be glad this exists once I build it, and what would they have done before mine showed up?
  • How could this idea be molded into something that becomes a game changer for the people it serves? What would the bigger version look like?
Advisor notes. Do not let students shoot down their own ideas in the early brainstorm, since ambitious ideas are exactly what you want at this stage. The role-simulation step is where the real growth happens, because students see the difference between an idea they like and an idea other people would actually want. They start to step into the role of a leader and a visionary by sitting in the seat of the people who will use what they build. Common project types include websites, apps, games, short films, business plans, community tools, scripts, art generators, educational tools, songs, and newsletters. Keep the list of ideas, since you will need it again later.
Flexibility. The brainstorm and napkin pitch work fine with pen and paper, and the AI focus group needs only a chat AI. If your school restricts AI access for students, the advisor can run the focus group on a shared screen and let students drive the questions.
Week 6

Scope, audience, scale, and the honest cost of building.

Goal: every student leaves with an approved, achievable project, a clear sense of who it is for and at what scale, and a real-world view of what their work would actually cost to build and run if it ever lived outside the classroom.

  • 0–5Define “shipped”
  • 5–25Scoping worksheet
  • 25–40Audience & scale
  • 40–60Cost reality check
  • 60–65Confirm
What happens
  • Advisor defines what “shipped” really means in this program. It is something a person outside the room can see, use, or read. It works, and the student is the one who built it.
  • Students complete the project scoping worksheet, which covers the project name, who it is for, what done actually looks like, what AI tools they will use, and their three biggest open questions.
  • Audience and scale conversation. Each student answers, in plain language, who they are building for, how many people that audience could reasonably include, and what changes about the project at different scales. A tool for one classroom looks very different from a tool for an entire school district, and naming the scale honestly makes the build week ahead much sharper.
  • Cost reality check. Even though no student is selling anything during the program, every student should understand the real cost of building something that lives in the world. Students use the AI to walk through the cost items their project would incur if it were real, including hosting fees, AI usage costs, design or art assets, ongoing maintenance time, marketing and reach, and any cost of the materials a hardware project would need. If a student would ever want to sell their thing later, they leave today knowing what their cost to make one and to deliver one would be, and what a fair price might look like.
  • Advisor circulates for quick one-to-one check-ins on whether the scoped project is genuinely achievable in five build sessions, and if it is not, the two of them scope it down together right there.
  • Every student leaves the room with a confirmed, written project and a one-page understanding of the audience, scale, and honest cost behind it.
Four scoping questions to ask every student
  • What does the working version look like? Describe it as specifically as you can.
  • Who is the first person you would show this to once it is real, and why them?
  • What is the one part you are most unsure how to build, and what is your first guess at the answer?
  • Which AI tool makes the most sense for what you are making, and what does it cost or include for free?
Advisor notes. Watch for two predictable failure modes: too big (an app that tries to do everything) and too small (nothing the student would be proud of). Push both groups firmly toward the middle. For students who cannot decide between two ideas, ask, “which one would you be most embarrassed not to finish?” The cost conversation is intentionally grounding, since it gives students a working professional’s view of what shipping something actually costs in time, money, and attention.
Flexibility. The worksheet can live on paper, in a shared doc, or as a simple template in any editor your school supports. The cost conversation works just as well with any chat AI and a calculator.
B Build Weeks 7–10

Make it real.

Four consecutive sprints. Something working beats something perfect. By week 10, every project is feature-complete and ready to polish.

Week 7

First working version.

Goal: every student has something running by the end of the session, however rough that first version may look.

  • 0–5Sprint rules
  • 5–40Build
  • 40–45Stand-up
What happens
  • Advisor opens: "Today's rule is 'working beats beautiful.' We want something on screen, not something perfect."
  • Thirty-five-minute build sprint. Students work on their projects using AI. Advisor circulates and helps unblock.
  • Stand-up. Each student says one sentence about what they built and what they plan to do next.
Advisor notes. The first build session is the hardest. Students often get stuck because they're trying to build the finished thing, not a rough version. Remind them to start with the core thing. If a student is completely stuck, help them break the project into one smaller step and do that step together.
Flexibility. This is where tech matters. For Chromebooks without dedicated coding tools, students can build with chat AI and publish using Google Sites, Canva, Google Slides, or Google Docs. Apps and games may require Replit or a similar tool, so confirm with IT before this session.
Week 8

Build sprint 2, adding features.

Goal: Core features take shape. Students build with purpose, not exploration.

  • 0–5Set goal
  • 5–38Build
  • 38–45Pair share
What happens
  • Before building, each student states what they will finish today. Write goals on the board.
  • Build. Advisor circulates.
  • Pairs share progress: "Here's what I got done, here's what I'm doing next."
Advisor notes. Goal-setting before the sprint matters more than it sounds. Students who name their goal are more likely to finish it. Watch for scope creep, since students who keep adding features before the core works. Redirect: "Does the thing you promised in week 6 work yet?" Encourage students to spend time on their project between sessions, even just thirty minutes, since cohorts that engage at home consistently ship more confident work.
Flexibility. If students need to collaborate or share screens and your environment restricts that, pairs can simply sit side by side and talk. No special tools required.
Week 9

Guide feedback (or peer review).

Goal: Students get outside perspective on their in-progress project.

  • 0–5Frame
  • 5–40Reviews
  • 40–45Next steps
Option A, with a guide available
  • Invite a returning guide (or a new one) to review projects.
  • Each student gets two to three minutes to show their project.
  • Guide asks three questions: What does this do? Who's it for? What's the hardest part left?
  • Advisor captures notes for each student.
Option B, without a guide
  • Structured peer review. Pairs do two-minute demos of their project.
  • Reviewer gives one thing working well and one question they have.
  • Swap and repeat with a different partner.
Advisor notes. This is the midpoint. Anyone way behind on scope: this is the moment to have an honest conversation about what's achievable by week 11. A 50% done honest project is more valuable than a 90% done project that lost its direction.
Flexibility. Guides can join remotely if in-person isn't possible. A 30-minute video call works. Peer review needs no technology at all.
Week 10

Feature complete.

Goal: Core functionality is done. Polish begins.

  • 0–5Define complete
  • 5–35Final push
  • 35–451:1 check-ins
What happens
  • Advisor opens: "Feature complete means the main thing works. We are not adding new things, we are finishing what we said we would build."
  • Final focused build session.
  • Advisor does quick 1:1 check-ins: what's done, what's left, is the project ready to show on week 12?
Advisor notes. Change the language: shift from "building" to "finishing." It signals that the phase has changed. If a student's project isn't working yet, scope down ruthlessly to get something working. A simple, working version is a win. An ambitious, broken version is not.
Flexibility. Whatever tools students used to build, they continue with. No change needed.
E Emerge Weeks 11–12

Ship and share.

Every student presents their work publicly. Families, administrators, and community see what got built. The showcase is the proof, and it is the beginning of a portfolio.

Week 11

Showcase prep.

Goal: Every student can present their project clearly and confidently in three minutes.

  • 0–10Format + rubric
  • 10–30Write + practice
  • 30–45Pair practice
What happens
  • Advisor explains the showcase: three minutes, show the thing, tell the story. Families, administrators, and community are invited.
  • Students draft their presentation using the framework below.
  • Pairs practice with feedback. Two questions: was it clear what they built? Did they own it?
Three-minute presentation framework
  • What did you build? Show it. (20 seconds)
  • Who is it for and why does it matter? (30 seconds)
  • How did AI help you make it? Be specific. (30 seconds)
  • What was the hardest part, and how did you figure it out? (30 seconds)
  • What would you build next? (20 seconds)
Advisor notes. Students often undersell themselves. Coach them to say "I built this" rather than "the AI made this," because the AI was the collaborator and they were the director. Let nervous students practice multiple times. The goal is comfort, not perfection.
Flexibility. Students can present live, with slides, a video, or just their working project. Whatever lets them show the work clearly in three minutes.
Week 12

Showcase.

Goal: Every student presents publicly. Families and community celebrate the work.

Format
  • Invite families, administrators, guides, sponsors, and local press two weeks out.
  • Set up project stations or run a presentation format, whichever fits your space.
  • Each student presents for three minutes.
  • Advisor opens with a brief intro to the program.
  • Optional: invite one guide to give brief closing remarks.
What to collect
  • Photos of students presenting (only with parental consent on file).
  • Project URLs, screenshots, or short video clips for portfolio use.
  • A guide endorsement note for each student who made a meaningful connection.
  • Advisor fills out the end-of-semester review the week after.
Advisor notes. Celebrate everything. First-time builders shipping any working project is remarkable. Have a plan for tech issues: every student should have a backup (screenshots or a short recording) in case a live demo fails.
Flexibility. The showcase can be in-person, streamed, or hybrid depending on what your school supports. A gym, library, cafeteria, or classroom all work.
Next steps

After the showcase.

Capture what worked, pass it forward, and start the next cohort stronger.

End-of-semester review

A one-page template. What worked, what to change, what the next advisor should know.

Student portfolios

Share project links, screenshots, and guide endorsements. Students own their work forever.

Start the next cohort

You now have a running program. Pass the framework to a second advisor or run it again.

Questions, feedback, or running into something this curriculum doesn't cover? Reach out at [email protected]. We respond to every email and improve the curriculum based on what advisors tell us.