SuperBrain AI Incubator Educational Methodology White Paper

Let passion create the future—in the real world.

SuperBrain AI Incubator is a decentralized, nonprofit, global innovation education experiment. Since its launch in 2023, we have hosted six hackathons, five online Human-AI Symbiosis Challenges, one AI for Good Winter Camp, and are currently incubating three projects—serving over 200 learners across Beijing, Shanghai, Guangzhou, Shenzhen, Chengdu, Dali, Hong Kong, Australia, and New Zealand.

This white paper systematically presents the educational methodology framework forged through SuperBrain's four years of practice—from era-level insights to educational goals, from core principles to operational methods, from product formats to hands-on experience, and from lessons learned the hard way to practices rigorously validated. We aim to offer parents and education professionals concerned with AI-era learning a substantive, evidence-based, and actionable professional reference.

Part I: Our Understanding of the Era

1.1 The Industrial-Era Education Ship Is Taking on Water

Today's education system originated centuries ago during the Industrial Revolution—designed to cultivate standardized 'executors' suited to assembly-line logic. Its core function is selection and refinement: training individuals to become compliant, interchangeable parts of society's machinery.

This logic is now failing—not incrementally, but systemically—in the AI era:

The screening value of academic credentials is collapsing. Artificial General Intelligence (AGI) may emerge within the next 5–10 years, displacing 90% of procedural, repetitive white-collar jobs; embodied AI robots will assume most blue-collar roles. A Harvard Business School report shows junior white-collar positions are being rapidly automated by AI—even graduates from Stanford, Tsinghua, Peking, Fudan, and Shanghai Jiao Tong face mounting employment challenges. Large-scale layoffs, beginning in the internet sector, are spreading across traditional industries—and from Silicon Valley outward, globally.

A vivid analogy: After 12 years of rigorous study—passing 'Driving Test Parts I, II, and III' repeatedly—you finally earn your license—only for AI to inform you that the future is fully autonomous.

Knowledge barriers are dissolving. In the face of AGI, the value of knowledge accumulated by any individual plummets. If education remains centered on knowledge transmission, it's like building castles on sand.

Career pathways are fracturing. In the industrial era, employment opportunities were highly centralized—90% worked for companies, and education served primarily as a filtering mechanism. In contrast, the AI era will foster a decentralized professional ecosystem, where an estimated 90% will become freelancers or 'slash' professionals. The linear path of 'school → job search → promotion' is no longer reliable.

1.2 A Fundamental Shift in Talent Demand

The source of commercial value is shifting. In the industrial era, commercial value stemmed mainly from material production and consumption. In the AI era, material abundance will be extreme—per Sam Altman's 'Everything Moore's Law,' costs of products and services halve every two years. Commercial value will increasingly derive from emotional resonance and creative experience.

From 'Tool Person' to 'Creator'. The industrial era demanded 'tool people' and 'cogs'—valuing execution and standardized problem-solving. The AI era calls for radically different roles—where core competencies shift toward curiosity, imagination, creativity, aesthetic sensibility, and interpersonal connection.

Two Core Roles in the AI Era. Society will bifurcate into two categories of roles AI struggles to replicate: First, those who leverage powerful curiosity, imagination, and problem-definition skills—using AI to become 'Super Individuals'—such as entrepreneurs, designers, and scientists—who operate as 'one person, one team.' Second, those who deliver uniquely human warmth and emotional support—'Life Companions'—such as coaches, healers, and artists—whose work AI cannot authentically replicate.

1.3 SuperBrain's Response: Not Repairing the Old Ship—Building a New One

Faced with this paradigm-shifting transformation, SuperBrain does not pursue incremental upgrades to traditional education ('Education + AI'), but instead builds an AI-native educational environment ('AI × Education').

SuperBrain doesn't teach children how to use AI tools—we immerse them in solving real-world problems, helping them learn to coexist and co-create with AI, transforming inner drive into tangible creativity. Here, children are no longer passive consumers of education—they become definers and creators of the new era.

When academic credentials can no longer anchor life's meaning, young people need new coordinates of purpose. SuperBrain's answer is: passion and creation. Empower children to learn and create—with passion—in the real world. That is SuperBrain's reason for being.

Part II: Three Core Educational Goals

What kind of people should we nurture in the AI era? Based on four years of practice, SuperBrain identifies three interwoven dimensions—not three distinct types of people—but essential capacities every young person should develop. Their integrated expression is what we call the 'Super Individual'.

2.1 AI Native

Collaborates fluently with AI, experiencing it as a natural extension of thought—not as an external tool.

While AI technology itself may take 5–8 years to mature and stabilize, the human cognitive shift must begin much earlier. Today's children are the AI era's 'natives'—they intuitively integrate AI into their thinking without friction. SuperBrain's mission isn't to wait for AI technology to mature before starting education—but to ensure that when the future arrives, a cohort of young people already possesses the mindset and creativity to master it.

Core Competencies: Human-AI symbiotic thinking, mastery of AI tools, and metacognition (knowing when to rely on AI—and when human judgment is essential).

Evidence from Practice: – In the 2024 Hackathon, a middle-school team with zero programming background used Claude + Unity to build the AI Native multi-agent game 'Grandpa's Egg'—NPCs integrated LLM APIs for dynamic dialogue, where conversation memory shaped pet personalities at birth. – Three middle-school students progressed from zero Apple development experience to shipping a Vision Pro app, 'Realmefy,' within one week; after two months of continued incubation post-hackathon, it launched on the App Store. – A nine-year-old learner completed a full commercial loop—from original IP design to blind-box product—using AI.

Together, these cases illustrate: Once AI removes technical barriers, imagination and initiative become the true bottlenecks.

2.2 The VUCA Player (Explorer of the Unknown)

The courage and capacity to find direction amid uncertainty.

VUCA—Volatility, Uncertainty, Complexity, and Ambiguity—is not an obstacle to eliminate, but the new normal in the AI era. Rather than seeking certainty, the VUCA Player maintains emotional stability, adapts rapidly, makes decisions amid ambiguity, and spirals upward through iterative experimentation and learning from failure.

Core Competencies: Antifragility, rapid adaptation, decision-making under ambiguity, emotional regulation, and learning from failure.

Evidence in Practice: - The carefully designed seven-day emotional arc of the Winter Camp itself constitutes a deliberate VUCA experience—students begin Day 1 saying, "I don't know what to do," navigate team conflict, technical roadblocks, and complete project overhauls, and by Day 7 confidently declare on stage, "I did it." - The Coach Handbook documents a highly introverted girl who maintained steady output throughout the hackathon—"completely unaffected by her teammates' emotional states"—and ultimately became her team's anchor. - The StrideVerse project underwent two full creative pivots; yet because the team shared a common visual prototype, each pivot felt like an "improvement," not a "failure."

Antifragility is not innate talent—it's an experience that can be intentionally designed. SuperBrain deliberately engineers "Mess-Up Nights" into its curriculum—not aiming for guaranteed outcomes, but committing instead to rapid reflection and iteration. Michael, founder of the Winter Camp, explained this in a lengthy message to parents on Day 3: it's a purposeful enactment of the VUCA Player educational objective.

2.3 The Unique Creator

Valuing one's uniqueness and realizing original creation rooted in personal life experience.

A Unique Creator is not synonymous with technical mastery. At its core lies the question: "Why do I create?"—the most powerful creations emerge from authentic, lived experience. Your pain, curiosity, and passion are irreplaceable raw material for creation—and precisely what AI cannot replicate.

Core Competencies: Self-awareness, original expression, transforming personal experience into meaningful creation, and problem-framing ability.

Evidence in Practice: - A high school student developed an early Alzheimer's screening game based on her grandfather's illness (Phase 1.0, 2023). - Xiao Bao (age 9) designed a blind-box product inspired by his own parent-child relationship, completing the full commercial loop—from design and production to market sales. - Middle-schoolers from the Menghe Campus team built an AI-powered virtual campus to give peers a safe outlet for emotional release.

The starting point for these projects was never "a teacher-assigned assignment," but rather "I truly care about this." ".

2.4 The Super Individual: Integration of the Three Goals

The integrated embodiment of the three developmental goals is the Super Individual—equipped with AI Native tool fluency, VUCA Player adaptability, and Unique Creator originality, capable of defining their own value in the AI era.

We don't train people who 'use AI.' We cultivate Super Individuals equipped with the core competencies essential for thriving in the AI era. This isn't a distant vision—it's the goal actively realized in every SuperBrain activity.

Part III: The Methodological Framework

If the developmental goals answer 'What kind of person do we aim to cultivate?,' the methodological framework answers 'How—and by what means—do we cultivate them?' SuperBrain's methodology comprises three pillars, one learning cycle, three product formats, and a set of supporting mechanisms.

3.1 Pillar One: Real Creation

High-quality learning happens in authentic contexts.

This is SuperBrain's most fundamental distinction from all PBL and STEM education providers. SuperBrain projects are not simulations or competition entries—they are acts of creation with real users, genuine market feedback, and tangible commercial viability.

Key Difference from Traditional PBL:

Most other educational institutions' projects lack commercial closure—they exist only as showcase pieces or contest submissions. SuperBrain projects pursue commercial sustainability: middle-schoolers build prototypes, university students develop products, and startup mentors guide commercialization. Students share in project revenue. Strong projects may enter incubators—and upon achieving Product-Market Fit (PMF), even launch full-time teams. This isn't 'pretend entrepreneurship'—it's authentic value creation.

A pivotal moment recorded in the Coach Handbook: after completing the full journey from confusion to creation, a teenager said:

"I made this thing—so I feel accomplished."

The Coach Handbook comments: "That sentence is the hackathon's most important outcome—not the app, not the presentation—but a young person's heartfelt affirmation: 'I did it.'."

The transformative power of Real Creation lies in giving children the confidence to face an uncertain future—not confidence drawn from accumulated knowledge, but from the lived experience of having 'made something real.'

3.2 Pillar Two: Life Touches Life

Coaches are not 'knowledge transmitters' or 'skill trainers,' but 'life companions.' This distinction isn't one of degree—it's one of dimension.

True education is life touching life. Under this philosophy, AI is merely a tool, and entrepreneurship serves only as a vehicle for 'cultivating character through action.' The long-term impact of entrepreneurial mentors' values, worldview, and life presence on young people runs far deeper—and lasts far longer—than knowledge transfer alone.

SuperBrain believes the most exceptional educators often operate outside traditional school systems—in the front lines of innovation and entrepreneurship.

The 'See · Respect · Act' Framework

The core philosophical framework in the SuperBrain Coach Handbook is 'See · Respect · Act':

• See: First, see the child as a whole person—their passions, fears, and inner state—not just their 'problems' or 'potential.' Seeing requires 'angel glasses': when students display 'problem behaviors' (e.g., phone obsession, passive resistance), look beneath to uncover positive underlying motivations. For example, a student anxious about mosquitoes may possess strong user empathy; a student playing with plastic bags may demonstrate material or tactile sensitivity.
• Respect: Honor each child's pace, choices, and uniqueness—never judge diverse lives against a single standard. The Coach Handbook emphasizes linguistic nuance: coaches never call students 'kids,' but always 'youth' or 'classmates,' because language itself conveys power dynamics.
• Act: Grounded in seeing and respecting, guide students toward autonomous action through thoughtful questioning, constructive challenge, and timely support. Intervention timing follows the principle 'Do not enlighten unless the student is eager to learn; do not prompt unless they are struggling to express'—teaching only when students are genuinely stuck, not before.

Practice of Self-Determination Theory (SDT)

This framework rests on Deci & Ryan's Self-Determination Theory—fulfilling three fundamental psychological needs: Autonomy: 'This is my choice'—honor students' project selections, even if coaches personally disagree. Competence: 'I can do this'—break tasks into 'just-right' challenges where small wins are visible and accumulable. Relatedness: 'I belong here'—interpret rule-challenging behaviors (e.g., late-night work, extra snacks) as expressions of belonging needs, and channel them into team rituals—not suppression.

Case Study: Xiao Y's Shadow Puppet Choice

During the Winter Camp, Xiao Y faced a choice between a psychology project and a shadow-puppetry project. Coaches viewed puppetry as failing to address real-world problems—but Xiao Y insisted. Coach Xiao Xiao didn't correct her choice; instead, he built trust through informal conversations—like sharing ice cream—until the true motivation surfaced: Xiao Y's friend's classmate was experiencing a mental health crisis, and she wanted to help but felt powerless. Shadow puppetry became her safe, expressive vessel for the issue she deeply cared about—adolescent mental wellness.

Coaches ignite not skills—but will.

The Coach Handbook's central thesis crystallizes in a parent's remark at a sharing session: 'Willpower outweighs capability.' Traditional education assumes capability must precede action—but in the AI era, that sequence flips: will comes first, and capability follows.

SuperBrain's Coach Standards

A SuperBrain coach does not need to be an AI expert—but must possess: - Real entrepreneurial or creative experience—grounded in the confidence that comes from having 'actually built something' - Genuine curiosity about children—not 'I will teach you,' but 'I want to understand you' - Alignment with—and willingness to practice—the SuperBrain coaching philosophy - The capacity to maintain emotional and psychological stability even under high-intensity conditions

The Coach Handbook uses a precise metaphor: The coach is a forest guardian—observing the entire ecosystem, not pruning each tree. It demands 'seeing both the trees and the forest': noticing the unique texture of an individual student's bark, while also sensing how sunlight and wind flow across the whole forest.

3.3 Pillar Three: Meaning-Making

When children's creations connect directly to real-world social challenges, their intrinsic motivation and quality of output undergo a qualitative leap.

This is the 'newest methodological insight' from SuperBrain's three-year experiment. From open-ended exploration in 2023, to themed tracks in 2024, to the targeted 'AI for Good' focus in 2026—each shift toward deeper meaning has consistently elevated student engagement and output quality.

Meaning-making isn't valuable simply because 'doing good' is noble—rather, meaning is humanity's deepest source of motivation—transcending external rewards, and even surpassing interest itself. When a student knows they're building an app to help blind children run, their drive to work past midnight comes from an entirely different place.

Real-Client Mechanism

The Winter Camp introduced the 'Real Client' mechanism—where project requirements originate from actual people with disabilities or community organizations. Clients—including Song Haifeng (founder of Yi Shu Guang, a blind-running initiative), Jenna (accessibility expert), Gao Shan, and Guan Qin (co-founders of an accessibility incubator)—participated throughout, rotating among project teams to answer questions and ensure student solutions genuinely addressed real needs.

The head of Avenues School wrote in their Winter Camp observation notes:

'Project design centered on "real problems"—introducing real clients and authentic needs—gives student projects tangible social significance.' This is the key driver of students' intrinsic learning motivation. "

SuperBrain isn't just cultivating 'children who can use AI'—it's cultivating 'children who understand why they should use AI.'

3.4 The AI-Empowered 'Build–Learn–Reflect' Cycle

Traditional education follows the cycle: 'learn knowledge → do practice → take exams.' SuperBrain flips it: First, build something → learn while building → reflect to elevate experience into understanding. AI makes this inversion possible—by removing the barrier of 'you can't do it if you don't know how.'

Build (AI-Empowered Building)

Students treat AI as a co-creator, building real products from scratch. AI lowers technical barriers—so creativity and intent, not coding fluency, become the limiting factor.

During the hackathon, Eva tackled the technical challenge of character keyboard movement. Technical support provided a complex set of pre-written code—but Eva chose instead to implement her own solution from scratch, collaborating with Claude late into the night. Her final version was cleaner and more effective. What she learned wasn't 'how to code keyboard movement'—but 'I can tackle anything I don't yet understand, together with AI.'

Later, students became so absorbed debugging that they forgot to eat or rest. This deep immersion isn't forced—it's the natural flow state that creation itself delivers.

Learn (AI-Empowered Learning)

It's not 'learn first, then do'—but rather learn by doing, and learn when stuck. AI serves simultaneously as knowledge base and instant tutor—students ask AI questions anytime, receiving personalized, just-in-time learning pathways.

During the hackathon, students needed to master programming, AI tools, and project delivery—all within five days. Traditional instruction would split this into 'Day 1: Learn programming; Day 2: Learn AI; Day 3: Start building.' SuperBrain starts building on Day 1—and learning happens on-demand (just-in-time learning).

The pedagogical research behind Human-AI Co-Creation Challenges responds to concerns that 'AI weakens children's thinking skills': negative effects stem from improper usage—e.g., asking closed questions and stopping thinking once an answer appears. Switching to open-ended questions—and using AI to deepen inquiry—instead sparks curiosity. Time and mental energy saved are redirected toward higher-order thinking, resulting in genuine cognitive growth.

Reflect (AI-Empowered Reflection)

Finishing isn't the end. SuperBrain employs a systematic reflection framework:

Spiral Inquiry—its core tool is a deceptively simple question: " Given everything you now know, what is the most important thing that should exist—but doesn't yet?" The power of this question lies in its spiral ascent—each answer builds on the prior round's insights, digging one layer deeper. It applies seamlessly to project ideation, daily stand-ups, coach-led prompts, and student self-reflection.

Multi-Level Review—Daily review (team reflection sessions each evening during hackathons/camps), phase review (coach and operations debriefs after each program), and cross-cycle iteration (SuperBrain's evolution from v1.0 to v4.0—each version shaped by rigorous review of the prior one).

Growth Narrative—the pitch and closing session invites students to tell their personal growth story. This isn't just about showcasing project outcomes—it's transforming lived experience into narrative: turning 'what I went through' into 'what I realized.'

3.5 Product Architecture: A Growth Ladder from Experience to Creator

Each SuperBrain product format is more than just an 'activity structure'—it's a methodological vessel designed to fulfill distinct educational aims.

Stage One: The Novice Village (AI Clubhouse)

Zero-barrier entry—designed to give children the visceral feeling of 'I can build this' in the shortest possible time.

Targeted at grades 4–8 with zero AI background, delivered as a mini-hackathon (1–3 days), with highly scaffolded design ensuring every participant produces something tangible. The goal isn't to 'teach AI skills'—but to cultivate self-efficacy: dissolving the fear that 'AI is too hard' and igniting lasting curiosity.

In the Winter Camp's overall success-metrics pyramid, 'The feeling of "I can" + being seen' sits at the foundational 100% coverage layer—while 'a usable MVP' sits at the top. Not everyone will deliver a perfect product—but everyone will be ignited.

Stage Two: Hackathon

An intensive innovation sprint—experiencing the full cycle of 'identifying a problem → defining a solution → prototyping → pitching' within one week.

Designed for ages 12–17, the hackathon is the core vehicle for SuperBrain's pedagogy. Its in-person format (5–7 days) functions as an immersive bootcamp, emotionally driven and supported by a three-tiered assistance system. Its online format (8–12 weeks) offers deep, sustained practice—with real client needs and iterative development.

A central statement from the Winter Camp's master plan: 'What matters most in the in-person camp is delivering atmosphere—and a lifetime experience—a true rite of passage. Technical details may fade, but the ignited spark lasts far longer.' What we deliver isn't skill—it's the unshakable belief: 'I can.' "

Stage Three: Incubator (Venture Studio)

From prototype to real-world product—guided by startup mentors, taking hackathon projects into the market.

For standout teams emerging from hackathons, the incubation cycle runs 3–6 months, structured as a collaborative loop: 'middle/high schoolers build the prototype → university students develop the product → startup mentors handle commercialization.' Students share in project revenue—creating direct, tangible awareness of value creation. Once product-market fit (PMF) is achieved, full-time teams may form—with the incubator holding 10%, and students plus mentors collectively holding 10–20% equity.

Students aren't education consumers—they're value creators. This isn't rhetoric—the incubator's equity structure explicitly allocates shares to students, and commercial revenue flows directly to participating students and mentors.

3.6 Enabling Mechanisms

Gamified System Design

Drawing on Yu-kai Chou's Octalysis Framework (epic meaning, advancement, creativity, ownership, social influence, scarcity, unpredictability, loss aversion), SuperBrain implemented a points system, guild/team competitions, achievement badges, daily quests, and leaderboards. During the Winter Camp, the 'Super Individual Development Program' narrative framed each participant as a hero with unique attributes—progressing via star-rank tiers and seven-dimensional achievement badges.

Emotional Arc Design

The seven-day camp isn't a knowledge-delivery schedule—it's a carefully crafted emotional journey:

• Days 1–2: Empathy Awakening— "I can't" → "I see." Deep empathy is built through video sharing by people with disabilities, firsthand experience of the "One Beam of Light" blind-running initiative, and direct conversations with real-life case subjects.
• Days 3–5: Iterative Growth— "I can try" → "I'm making progress." Participants enter the project development phase—facing technical challenges, team conflicts, and strategic pivots—iterating continuously with coaching support and AI empowerment.
• Days 6–7: Narrative Expression— "I did it" → "I'll keep going." Through pitch presentations, growth storytelling, and pop-up booth exchanges, participants transform lived experience into enduring conviction.

Five Core Design Principles Guide Every Stage: Ignition before delivery; embodied experience before instruction; authentic feedback before polite praise; differentiated development before uniform effort; people over projects.

Embodied Experience First

Students learn by doing—not thinking. On Day 2 of the Winter Camp, an accessibility empathy exercise pairs students: one wears a blindfold while the other guides them—recreating daily realities for people with visual impairments. Afterward, student Chen Haotian shared: "I wanted to cry the moment I put on the blindfold." Only then did they begin their AI for Good projects—now, "building for users" carried entirely new weight.

Three-Tier Support System

Startup Mentors(Strategic Direction) +Coaches(Life Companioning) +Navigators / Teaching Assistants(Technical Support)—these roles must remain distinct. Coaches don't teach coding; mentors don't provide daily emotional support; navigators don't make strategic decisions. At the Winter Camp, Michael captured the spirit of this system in one line to the coaching team: "I'll take the blame—I'm handing full ownership over to you now." Avenues School's Head of School noted in her observation journal that this high-trust delegation powerfully ignited team autonomy and accountability.

Part IV: Four Years of Practice

SuperBrain's evolution isn't the story of a prewritten plan being executed—it's the unfolding journey of an educational philosophy colliding with reality, constantly adapting, refining, and evolving. Each stage builds directly on insights from the last; every 'version upgrade' emerges from deep reflection on the prior iteration.

4.1 SuperBrain 1.0 (2023): Experimental Validation—"Can students truly innovate?"

In early 2023, SuperBrain was still called 'Future Stars.' It launched as an experimental summer startup incubator—no business model, no full-time team, just a group of education-focused entrepreneurs willing to 'build for love.'

Over two months of incubation, startup mentors led small teams, invited heavyweight industry guests for internal talks, and enrolled students in AI training. The results were astonishing: one high schooler, inspired by his grandfather's Alzheimer's diagnosis, developed an early-screening game for the disease; numerous elementary and middle-schoolers delivered innovative projects far exceeding age-based expectations.

Core Validation: The pedagogical model of 'learning and creating in the real world' is viable. Shift in Assumption: From "serious innovation requires seasoned professionals" to "with appropriate guidance, young students can produce meaningful creations."

4.2 SuperBrain 2.0 (2024): Structured Hackathon—"Can AI empower absolute beginners to build real products?"

Having validated feasibility in 1.0, 2.0 focused on structure: compressing the two-month timeline into one month of online AI co-learning plus one week of intensive hackathon, launching four challenge tracks (Quadruped Robot Dogs, AI + Creative Culture, AI + Gaming, AR/VR), and introducing—for the first time—the three-tier support system (startup mentors + coaches + university student TAs). Two hackathons served over 20 participants.

Key Outcomes: — A middle-school team with zero programming background used Claude to build an AI Native multi-agent game in Unity. — The robot-dog team achieved collaborative interaction between large language models and hardware. — Elementary students in the creative culture track developed a 'Minecraft'-themed board game with AI assistance. — Three middle-schoolers on the VR team progressed from zero Apple experience to publishing a Vision Pro app on the App Store.

Core Validation: AI fundamentally reshapes 'teachability'—absolute beginners can achieve production-grade output via 'Vibe Coding' (co-programming with AI). Shift in Assumption: From "learning AI requires programming foundations" to "AI itself is the best learning scaffold." Methodological Milestone: Established the core framing: 'Learn and create in the real world through passion.'

4.3 SuperBrain 3.0 (2024–2025): Human-AI Symbiosis Challenge—"How should teachers and AI collaborate?"

Though 2.0 yielded strong results, post-mortems revealed two deeper issues: unclear AI role definition, and lack of methodology for teacher intervention and collaboration. This propelled SuperBrain into its more profound 3.0 exploration—the Human-AI Symbiosis Challenge.

—a fully online program releasing structured weekly tasks using the Challenge-Based Learning framework: Identify Problem → Form Team → Deep Research → Plan MVP → Execute → Showcase Outcome. Tasks launch live every Wednesday at 8 PM; pedagogical discussions follow at 9 PM.

Scale Far Exceeded Expectations: Anticipated 3–5 teams; instead, over 100 individuals registered, 48 teams joined, and 15 completed and presented. Participants ranged from Grade 1 students to adults, spanning Beijing, Shanghai, Guangzhou, Shenzhen, Hong Kong, Chengdu, Dali, Nanchang, Dongying, Australia, and New Zealand. Task documents received 3,000+ views from 500+ readers; 14 live sessions were held, with task-debrief replays watched over 1,000 times.

Five themed cycles focused progressively on 'Playfulness,' 'Making Family Life More Fun,' 'AI & Age-Friendly Design,' 'Scientific Exercise,' and 'Gamifying Sports'—each diving deeper into distinct social contexts. Twenty-two projects debuted at the 2050 Conference during Cycle 3.

Core Validation: The human-AI collaboration relationship *is* the curriculum—AI is not a teacher-replacement tool, but a mirror reflecting human collaboration patterns. Shift in Assumption: From "AI is a student productivity tool" to "how humans and AI collaborate is itself a core competency to be taught." Methodological Milestone: A replicable human-AI collaborative teaching model emerged—with all pedagogical outputs open-sourced.

4.4 SuperBrain 4.0 (2026): AI for Good Winter Camp + Incubator—"Can empathy + AI solve real-world social problems?"

4.0 synthesizes all prior methodology—integrating lessons learned into a focused 'AI for Good' theme, culminating in a 7-day, 6-night in-person Winter Camp held in February 2026.

Scale: 30 enrolled participants (27 completed) × 33 Navigators/Volunteers. Recruitment lasted only 37 days (target: 2–3 months); WeChat posts were shared 311 times; enrollment included 18 paying participants, 4 scholarship recipients, 6 discounted or gifted spots, and 5 returning SuperBrain alumni. Participant profile: 60% female, 60% aged 13–14 (with 42% aged 13—the 'golden age': self-aware yet unburdened by exam pressure); 80% had prior coding experience, 20% were absolute beginners.

Core Innovation: Introduced real-life case subjects—including visually impaired advocate Song Haifeng and accessibility expert Zhan Na—to guide students in building AI solutions for actual disability communities. Three incubated projects emerged: a feedback platform for accessibility issues ('Complaint Platform'), a blind-running volunteer-matching app ('Blind-Runner Assistant'), and speech recognition for children with articulation disorders (Project Resonance—already deployed live).

Post-Camp Incubation: ~40% of participants (~12 students) entered an 8–10-week incubation phase—transforming camp 'ignition' into 'deep output.' A points-based system was introduced (3 pts for technical contribution + 3 pts for collaboration + 4 pts for learning); participants scoring ≥8 points for three consecutive weeks advanced to 'Core Member' status, unlocking 1:1 mentor access and project decision rights.

Core Validation: When empathy experience + AI empowerment + real case subjects converge, student intrinsic motivation and output quality reach unprecedented levels. Shift in Assumption: From "learning AI is about skill acquisition" to "learning AI is a journey of solving problems *for others*."

Part V: Practical Insights from Three Years of Implementation

This is the whitepaper's most distinctive, value-driven section. While most AI-in-education whitepapers tout 'our philosophy' and 'our methodology,' few systematically document 'what pitfalls we encountered, which assumptions we revised, and which methods proved invalid.'

5.1 Assumptions We Revised

5.2 Methods Proven Effective

Environment Engineering > Process Management. Winter Camp Day 2 was the "mass collapse day"—simultaneous safety incidents, parent complaints, volunteer boundary violations, and operational breakdowns. The conventional response would be tighter process control. SuperBrain made a pivotal decision on Day 3: instead of trying to "control everything," it established a four-track management framework: (1) Teaching Track (aiming high), (2) Parent Track (maximizing satisfaction), (3) Camp Operations Track (ensuring baseline safety), and (4) Communications Track (adding polish). Result: overall stability rapidly recovered from Day 3 onward, culminating in exceptionally high parent satisfaction. Key insight: Teaching can be messy—but the learning environment must remain stable (safe, respectful, trustworthy).

"Not abandoning any child" in practice. During the Winter Camp, 13-year-old Xiao Yang (pseudonym; 185 cm tall, dropped out of Grade 4) requested to withdraw on Day 1. The team refused to give up. Instead, they gave him space and dedicated one-on-one time—he spoke with mentor Ms. Song for 2–3 hours. After receiving positive reinforcement for serving tea to guests, he opened up. He not only completed the camp but also designed an accessibility-focused experience product for volunteers. His mother said: "Spending over ¥10,000 was absolutely worth it!"

Returning students serve as the "ballast." Returning SuperBrain students comprised 18% of the Winter Camp cohort—they anchored the camp's floor: while newcomers were still adjusting, returning students were already fully engaged, lifting the collective energy.

Key Success Factors for Projects—Cross-project comparative analysis:

5.3 Methods Disproven or Requiring Adjustment

The "single-tool AI learning" model is ineffective. The 2025 Strategic Retrospective explicitly concluded: traditional "teaching AI tool usage" lacks a systemic framework, yields unpredictable outcomes, and delivers poor parent satisfaction. We must shift decisively to "learning AI on-demand within authentic projects."

Volunteers cannot begin work without training. All 33 Winter Camp volunteers started without systematic training, formal agreements, or orientation sessions. Day 2's collective collapse was inevitable. Though recovery accelerated after Day 3, this "fight first, train later" model carries unacceptably high risk. Lesson: At minimum: half-day alignment on philosophy + explicit role boundaries + emergency scenario drills.

Hardware projects require upfront technical validation. The Smart Medicine Cabinet project severely underestimated the difficulty of its WiFi module integration and 3D modeling—hardware implementation ultimately failed. Lesson: Any hardware-involved project must undergo POC (Proof of Concept) prior to the hackathon.

"Adult-curated ideas" fail to ignite student engagement. In the Menghe Campus project, the direction emerged from adult discussions (a Japanese-style campus management simulation game)—yet students had never played such games and lacked emotional resonance. Result: low group identification and "no sense of team." Lesson: Problems must either be discovered by students themselves—or at least resonate deeply with their lived experiences.

5.4 Practical Recommendations for Peers

Drawing on four years of experience, we've distilled the following actionable, field-tested recommendations:

1. Shift from 'Teaching AI' to 'Doing with AI'— Start building projects on Day One; learn just-in-time through doing—not by completing courses before practicing
2. Introduce Real Clients— Empower students to solve real problems for real people; meaning is the strongest engine of intrinsic motivation
3. Design Emotional Arcs, Not Syllabi— The core design dimension of a camp or course is the student's emotional journey, not a checklist of knowledge points
4. Distinguish Three Support Roles— Directional guidance, life mentoring, and technical support should never be conflated into a single role
5. Invert Success Criteria— Prioritize 'being ignited' as a 100% baseline requirement; treat 'polished final products' as aspirational—not mandatory
6. Pre-Schedule a 'Breakdown Day'— Day 2 of any multi-day camp is typically when challenges peak; proactively build in emotional release mechanisms
7. Strengthen Pre-Selection— Clarify whether the project idea originates from the student or is externally assigned; externally assigned ideas show significantly lower success rates
8. Mixed Ages Are a Feature, Not a Flaw— Leverage age diversity to spark role differentiation and peer-led leadership

Part VI: Evidence from Real Cases

The cases below are ordered chronologically by activity date. All participant names have been anonymized.

I. SuperBrain 1.0 Hackathon (2023)

Case 1: Early-Stage Alzheimer's Screening Game — 'I Want to Help Grandpa'

Activity: 2023 SuperBrain 1.0 Startup Incubation Hackathon

Background: A high school student, inspired by her grandfather's Alzheimer's diagnosis, wanted to build a game to support early screening. This motivation emerged entirely from lived experience—no external 'assignment' was involved.

Outcome: With mentorship from startup coaches and AI assistance, the team delivered a working prototype of an early-stage Alzheimer's screening game. This was SuperBrain's earliest validation that, with appropriate scaffolding, young learners can produce meaningful creative work.

Validated Methodology: Unique Creator (life-experience-driven) + Meaning Construction (doing something real for a loved one)

II. 2024 AI Hackathon

Case 2: Bowu Pavilion / Realmefy — High Schoolers with Zero Prior Experience Launch an App on the App Store

Activity: 2024 AI Hackathon

Background: A team of high school students—none with prior Apple development experience—taught themselves Unity, Xcode, and Swift in seven days and built Realmefy, a 3D personalized emotional space app for Apple Vision Pro. Realmefy transforms personal memories and emotions into immersive, explorable 3D worlds.

Outcome: After the hackathon, the team continued iterating for two months—and successfully launched Realmefy on the App Store. This exemplifies the 'Super Individual' concept: a few high schoolers, empowered by AI, accomplished what typically requires a full startup team.

Validated Methodology: AI Native (AI-enabled path from zero to App Store) + Authentic Creation (full product lifecycle)

Case 3: Bipedal Robot Dog Mindog — Giving Cold Machinery a Warm Soul

Activity: 2024 AI Hackathon

Background: Middle and high school students collaborated across age groups, using large language models to endow a robot dog with natural conversational ability and emotional expression. The project received collaborative support from China's National Humanoid Robotics Lab.

Outcome: In seven days, the robot dog evolved from a 'command-executing machine' into a companion capable of chatting, expressing emotion, and even 'coaxing affection'. It became a signature case cited repeatedly by Michael Crawford in his talks on 'the upper limits of middle-school creativity in the AI era'.

Validated Methodology: AI Native (LLM + hardware integration) + VUCA Player (uncertainty management through cross-age collaboration)

Case 4: Xiao Xiao and the Unity Girls — Middle Schoolers with Zero Coding Experience Build an AI-Native Game, Featured on Oriental TV

Activity: 2024 AI Hackathon

Background: Four middle school girls (Xiao Xiao, Eva, etc.—all pseudonyms), with absolutely no prior programming experience, used natural-language interaction with the Claude LLM alongside the Unity engine to build an AI-Native multi-agent game, 'Grandpa's Egg', in seven days. In-game NPCs integrate LLM APIs for open-ended dialogue; their memory of past conversations shapes pet personalities at birth—giving each NPC distinct personality traits and branching storylines. Players report: 'NPCs grow like real people.'

Outcome: The project was featured on Oriental TV's documentary series 'Future China', validating the feasibility of adolescent 'no-code' development of complex games. Team members reflected: 'We used to think coding takes years—but now we see that if you can articulate your needs clearly, AI can bring them to life.'

Post-Camp Continuation: After the hackathon, Xiao Xiao demonstrated exceptional initiative far beyond her age: she recruited university and high school students to join the team, appointing undergraduates as project managers while assuming the role of product owner to drive ongoing iteration. A seventh-grader with average academic performance—and previously low self-confidence—emerged as a confident leader capable of orchestrating cross-age collaboration.

Validated Methodology: AI Native (zero-to-AI-native game development) + VUCA Player (from self-doubt to team leadership) + Authentic Creation (a real product undergoing sustained iteration). This case also vividly illustrates the principle 'Willpower > Skill': you don't need to master everything first—you begin, and capability emerges through action.

Case 5: Team Conflict & the 'Company Agreement' — Hands-On Training in VUCA Competence

Activity: 2024 AI Hackathon

Background: The 'Singularity' team (building an AI-powered NPC escape-room game) comprised four strong-willed students who clashed over roles and collaboration. Some issued directives; others responded with passive resistance.

Process: Coaches intervened—not by mediating directly or assigning roles—but by guiding the four students to co-create a 'Company Agreement': a set of self-defined rules and accountability mechanisms, designed and enforced collectively. This process itself was a micro-scale organizational development experiment.

Results Team cohesion significantly improved after signing the "Company Agreement," culminating in a fully functional AI NPC escape-room game prototype.

Validated methodology VUCA Player (finding collaborative pathways amid conflict) + Real Creation (project pressure accelerating collaborative skill development)

III. AI Creative Camp (Shanghai, August 2024)

Case Study 6: Changwei Miao — A 9-year-old girl creates her own original IP using AI and becomes the top-selling vendor at the camp fair

Activity AI Creative Camp, August 2024 (Shanghai Institute of Innovation & Creativity; 3.5 days; 16 participants)

Background An'an (pseudonym), age 9, with zero prior drawing experience. Throughout the AI Creative Camp, she used AI image generation to create her original cat IP, "Changwei Miao," then produced physical merchandise—including stickers, notebooks, bags, and standees.

Outcome She earned ¥500 at the Way to AGI pop-up fair—the highest revenue among all vendors. Post-camp, she continues operating a social media account documenting her creative journey. This exemplifies the principle that "once AI removes technical barriers, creativity and initiative become the only bottlenecks."

Validated methodology AI Native (zero foundation → AI creation → physical product) + Unique Creator (original self-expression by a 9-year-old) + Real Creation (end-to-end commercial loop)

Case Study 7: Xiao Baozi — A 9-year-old girl transforms parent-child conflict into a blind-box product

Activity AI Creative Camp, August 2024

Background Hanhan (pseudonym), age 9. Her biggest daily struggle? "Not wanting to do homework"—a recurring point of tension with her mother, representing her most authentic lived pain point.

Outcome She transformed this pain point into creative opportunity—designing a blind-box product that uses gamified mechanics to help families negotiate daily routines. Each blind-box component was hand-sculpted from colored clay, completing a full commercial loop: design → production → market sale.

Validated methodology Unique Creator (translating real-life experience into tangible product) + Meaning-Making (solving a problem she genuinely cares about)

IV. AI Sports Gamification Camp (Shanghai, July 2025)

Case Study 8: Spinal Rehabilitation Game — Gamifying solutions for sedentary lifestyles

Activity AI Sports Gamification Camp, July 2025 (Caohejing Science Park, Shanghai; 6 days; elementary to middle school students, no prior coding experience)

Background Participants observed worsening spinal health among peers due to prolonged sitting—but traditional rehabilitation faces three barriers: "no time," "no motivation," and "habitual preference for digital games."

Outcome They designed an AI-powered, gamified spinal health training product—completing end-to-end workflow: research, design, programming, and physical interaction. By wrapping rehabilitation exercises in game mechanics (points, levels, challenges), they turned tedious routines into engaging, self-motivated play.

Validated methodology Meaning-Making (acting on real issues affecting one's community) + AI-Enabled Learn-Make-Think Cycle

V. AI Hackathon (Kunshan NCC Nomad Community, July 2025)

A 7-day, 6-night immersive hackathon for students aged 13–17. Of over 70 applications, 32 were selected—achieving near 1:1 student-to-mentor ratio.

Case Study 9: Cha Yu — Reintroducing youth to traditional tea culture through gaming

Activity AI Hackathon, July 2025

Background A team of three middle and high school students (ages 13–17) developed an immersive tea-culture game in seven days—guiding players through the full journey from tea cultivation to running a teahouse.

Outcome They delivered a demonstrable MVP prototype, accompanied by a gameplay demo video. Their research revealed that 83% of teens have limited knowledge of tea culture—so they reframed "knowledge dissemination" as "experience creation." This project illustrates how middle-schoolers can fuse traditional culture with modern game design using AI.

Validated methodology Real Creation (a complete game product) + AI Native (AI-assisted game development)

Case Study 10: I Square — An intergenerational team builds an AI-powered goal-management app

Activity AI Hackathon, July 2025

Background A five-person team spanning ages from elementary school to adulthood: elementary students handled frontend development; middle-schoolers built backend systems; high-schoolers led product design; university students managed project coordination; and adult mentors provided technical guidance.

Outcome They built a dual-AI-engine goal-management software offering gamified emotional companionship for users who "have goals but lack self-discipline." The IP character was co-created via AI-human collaboration and 3D-printed. The team remained intact post-hackathon, continuously iterating and launching the product.

Validated methodology Real Creation (extending from camp project to live product) + Life Influencing Life (mutual learning across generations)

Case Study 11: StrideVerse — Redefining direction twice—and still delivering successfully

Activity AI Hackathon, July 2025

Background A three-member team: Meng (captain of his school soccer team—introverted yet charismatic), Xu (passionate about running but struggles with consistency), and Jiang (a logic-puzzle enthusiast and speed-solving Rubik's Cube master). Xu raised the core pain point: "I love running—but staying consistent is hard. Can we make it more fun through gamification?"

Process– Day 2: Direction confirmed—team highly energized.\n– Day 3 evening: First pivot—technical approach proved unfeasible.\n– Day 4: Second pivot—using AI workflows for UI design proved overly complex, triggering heightened anxiety. The team decisively shifted to Scratch—a simpler platform—and caught up in one day.\n– Days 5–6: Iterative refinement and minor bug fixes.\n– Day 7 Demo Day: Full team engagement—Meng transformed from a quiet task-completer into a charismatic, confident presenter.

Why didn't two major pivots derail the team? Because the Lead Mentor introduced three foundational software design principles from Day One: Modularization, Visualization, and Task Decomposition The team maintained a shared visual prototype throughout. Each pivot wasn't "starting from scratch," but rather "implementing the same vision in a different way"—making failure feel like "design iteration," not "project failure."

Validation Methodology— Authentic Creation (born from personal pain points) + VUCA Player (delivered successfully after pivoting direction twice) + Reflective Iteration (each pivot built upon insights gained in the previous cycle)

Case Study 12: Xiao Y's Shadow Puppetry — How a Coach Sees Beyond Appearances to Uncover Genuine Motivation

Activity AI Hackathon, July 2025

Background Xiao Y (pseudonym), a 14-year-old girl, chose shadow puppetry over a psychology project. Coaches initially judged that shadow puppetry couldn't address real-world problems.

Process Coach Xiao Xiao didn't correct her choice. Instead, he built trust gradually through informal settings—like sharing ice cream. After several conversations, her true motivation surfaced: a classmate of her friend was experiencing a mental health crisis, and Xiao Y wanted to help—but felt powerless. Shadow puppetry became a safe, expressive vessel for her deeper concern: adolescent mental well-being.

Key Insight Had the coach 'corrected' Xiao Y's choice immediately, that deeper need would never have been revealed. 'Seeing' requires patience and trust—and trust flourishes only in informal, de-escalated, power-neutral spaces.

Validation Methodology See · Respect · Act + Coaches don't teach skills—they ignite intention

Case Study 13: Xiao Zhang's Sustained Creation — From 'Loves Gaming' to 'Gamifying School Life'

Activity AI Hackathon, July 2025 → Post-Camp Sustained Creation

Background Xiao Zhang (pseudonym), a junior high student in Wuxi. Like many peers, he loved video games and knew almost nothing about AI—his mother was deeply anxious. Rather than urging him to quit gaming, we asked: 'When you play games, do you ever think something could be designed better?' He replied: 'Yes! Many games are boring—I think I could do better.' We said: 'Then go ahead and build one.' "

Hackathon Phase He and his team built an intelligent medicine cabinet, handling frontend development (a skill he'd never used before). He treated AI as his 'programmer', and himself as the 'product manager'. Though the hardware component remained incomplete, those seven days marked his first taste of genuine creative joy.

Post-Camp Momentum This is where the case truly shines—

1. Homework Management System Back at school, he identified pain points in homework management and built a web-based solution with a points system and leaderboard. Now used by his entire class—and spreading to others.
2. 'Dibang Exchange' A unified platform combining secondhand trading, lost-and-found services, and charity sales—with proceeds directly funding library books. A middle-schooler designed a fully functional commercial loop.
3. 'WordMaster' His most astonishing work—a gamified English learning platform featuring level-based challenges, head-to-head matches, seasonal leaderboards, and team formations. He translated his deep understanding of game design into a powerful learning tool. The school's IT department has now proactively provided him server access—and next month, he'll launch the product at a formal event in the school auditorium.

His Mother's Observation: 'He's shifted from being someone who plays games to someone who thinks in game mechanics—from solving problems to designing them.'

His Own Reflection: 'I realized creation doesn't require grand gestures—it starts with small things around me, solving real problems right where I am.'

Validation Methodology A full-spectrum validation of the SuperBrain methodology—AI Native (AI-enabled development from zero coding experience) + Unique Creator (channeling passion for gaming into product innovation) + Authentic Creation (solving tangible, local problems) + Reflective Iteration (spiraling upward across successive products). A single 7-day hackathon didn't just teach skills—it ignited the belief: 'I can create.'

VI. AI for Good Winter Camp (Shanghai, February 2026)

Case Study 14: Xiao Yang's Transformation — 'Not Giving Up on Any Child'

Activity AI for Good Winter Camp, February 2026 (Caohejing Technology Park, Shanghai)

Background Xiao Yang (pseudonym), age 13, stands 185 cm tall and dropped out of Grade 4. On Day 1, he requested to withdraw from camp.

Process The team neither granted his withdrawal nor forced him to stay—instead, they gave him space. That afternoon, he spoke with mentor Mr. Song Haifeng (founder of 'A Ray of Light'—a blind-running initiative) for two to three hours. Afterwards, he voluntarily served tea to guests; receiving positive feedback, he gradually opened up. He then participated fully—and co-created an accessibility experience tool for volunteers.

Key Insight What Xiao Yang needed wasn't 'being taught'—it was 'being needed'. When he sensed his presence was seen and valued, intrinsic motivation emerged naturally. His mother remarked: 'Spending over ¥10,000 on this was absolutely worth it!'

Validation Methodology Life Impacts Life (See · Respect · Act) + Meaning-Making (doing work that matters for real people)

Case Study 15: Resonance / Project Resonance — AI Speech Recognition for People with Articulation Disorders

Activity AI for Good Winter Camp, February 2026 → Post-Camp Incubation

Background One of three incubated projects born during the winter camp. The team developed a specialized AI speech recognition tool for people with articulation disorders (whose speech is unclear and unrecognized by standard voice assistants), integrating voice cloning technology to help them 'be heard'.

Outcome This project advanced furthest among all winter camp initiatives—from prototype to sustained post-camp development, culminating in a full-stack React + TypeScript application that was deployed live(project-resonance.cn). It was selected as a core incubation project, and team members received dedicated one-on-one mentorship.

Validation Methodology Meaning-Making (deep motivation drawn from serving a real, underserved community) + AI-Powered Learn-Build-Think Cycle (a complete product journey from prototype to production) + Authentic Creation (a live, deployed solution)

VII. Human-AI Coexistence Challenge (Online-only, launched October 2024)

Case Study 16: 48 Teams Going Fully Online — Participation Exceeding Expectations by 10x

Activity Human-AI Coexistence Challenge (launched October 2024; five cycles; fully online)

Background Originally planned for 3–5 teams, the challenge attracted over 100 individual sign-ups and 48 teams. Participants ranged from Grade 1 students to adults—and spanned Beijing, Shanghai, Guangzhou, Shenzhen, Hong Kong, Chengdu, Dali, Nanchang, Dongying, Australia, and New Zealand.

Process• Weekly live task releases every Wednesday at 8 PM, following a structured, challenge-based learning process: Identify a problem → Conduct in-depth research → Plan an MVP → Implement → Present. Students apply systematic thinking tools—including Robert Crawford's Attribute Listing Method, Stakeholder Mapping, Six Thinking Hats, and SCAMPER—augmented by AI assistance. Task documentation has been viewed over 3,000 times by 500+ people; 14 live sessions have been held, with over 1,000 replays of the task debrief livestreams.

Outcomes• 15 teams completed and showcased diverse projects—from chemistry-learning board games and low-code herbal medicine tracking apps to Coze-powered vocabulary assistants. In Phase III, 22 projects were presented at the 2050 Conference.

Key Pedagogical Insights• Why spend hours brainstorming team names? Because it's a 'highly forgiving practice field': Forgetting brainstorming principles while naming a team only yields an unremarkable name—but failing to apply them when identifying problems or designing solutions creates real challenges. • How students use AI makes all the difference: Closed-ended questions shut down thinking, while open-ended questions spark curiosity.

Validated Methodology• AI-Augmented 'Make-Learn-Think' Cycle + Reflective Debrief & Iteration + Replicable Methodology Outputs

Comprehensive Case Index

SuperBrain has produced 41 student project cases over four years; the above 15 represent key examples. Full case coverage spans six activity tracks:

Methodological Framework Overview

┌────────────────────────────────────────────────────────────┐
│                                                            │
│   Core Philosophy: Empower children to learn and create in the real world—driven by passion. │
│                                                            │
├────────────────────────────────────────────────────────────┤
│                                                            │
│   Three Developmental Goals                                 │
│   ┌────────────┐ ┌────────────┐ ┌────────────┐            │
│   │ AI Native  │ │ VUCA Player│ │ Unique Creator│ → Super Individual │
│   └────────────┘ └────────────┘ └────────────┘            │
│                                                            │
│   Three Methodological Pillars                              │
│   ┌────────────┐ ┌────────────┐ ┌────────────┐            │
│   │ Real Creation │ │ Life Impact │ │ Meaning-Making │            │
│   └────────────┘ └────────────┘ └────────────┘            │
│                                                            │
│   Learning Loop       Product Ladder        Supporting Mechanisms │
│   ┌──────────┐   ┌──────────────┐  ┌──────────────┐      │
│   │ AI-Augmented │   │ ① Novice Zone │  │ Gamified System │      │
│   │ Make-Learn-Think │   │ ② Hackathon   │  │ Emotional Arc   │      │
│   │ Reflect & Iterate │   │ ③ Incubator   │  │ Embodied Experience │      │
│   └──────────┘   └──────────────┘  └──────────────┘      │
│                                                            │
└────────────────────────────────────────────────────────────┘

Appendix: Key Reference Documents