Deep Research Prompts v2 — GenEvolve Education Platform
Updated: Feb 28, 2026 Purpose: Prompts for GUI deep research tools (Gemini Deep Research, ChatGPT Deep Research, Perplexity Pro) Context: Building an open-source, self-hosted education platform for a UK "Education Village" — physical school + hybrid/online tiers, ages 4-18, child-led pedagogy based on Bloom's taxonomy, sovereign data (no Google/Microsoft dependencies), with alternative provision (AP) and SEND cohorts alongside neurotypical students.
ORIGINAL PROMPTS (v1)
PROMPT 1: Open-Source Education Platform Architecture & Stack Comparison
I'm building a self-hosted education platform for a UK "Education Village" school that serves ages 4-18, combining physical, hybrid, and fully online learning tiers. The platform must be:
- Open source (MIT/Apache/GPL), self-hostable on our own infrastructure (Docker/VM)
- Zero dependency on Google Education, Microsoft 365, or any Big Tech cloud (ICO Children's Code compliance, UK data sovereignty)
- Support child-led learning (students choose pathways, not prescribed routes)
- Bloom's taxonomy embedded in curriculum structure and assessment
- Multi-tenant (multiple village sites, eventually licensable globally)
- AI-native but model-agnostic (swap between providers via env vars)
Please do a deep technical comparison of these open-source education platforms as fork candidates:
1. LearnHouse (Next.js 14 + FastAPI + PostgreSQL + Redis) — MIT license
2. Canvas LMS by Instructure (Ruby on Rails + PostgreSQL) — AGPL
3. Open edX (Django + MongoDB + MySQL) — AGPL
4. Chamilo (PHP + MariaDB) — GPL
5. Moodle (PHP + MySQL/PostgreSQL) — GPL
6. ClassCloud (if open source — verify)
7. OpenOlat (Java + PostgreSQL) — Apache 2.0
For each, analyze:
- License implications for a commercial fork (can we white-label and sell?)
- Codebase size, language, framework maturity
- Built-in features: course authoring, assessments, portfolios, progress tracking, parent portal, community/discussion
- AI integration (existing or how hard to add)
- Multi-tenancy support (built-in or requires modification)
- Mobile experience (responsive, PWA, native app)
- Docker deployment readiness
- Activity and community health (GitHub stars, recent commits, contributors)
- Accessibility (WCAG 2.1 AA compliance for neurodiverse learners)
Produce a comparison matrix and a clear recommendation for which platform to fork, with reasoning. Consider that we have AI coding agents that can modify any codebase rapidly (4-6 week MVP timeline).
PROMPT 2: Child-Led Learning UX — How Bloom's Taxonomy Works in Software
I'm designing a learning platform where education is child-led — students choose their learning pathways rather than following a prescribed curriculum. The pedagogical foundation is Bloom's Taxonomy (Remember → Understand → Apply → Analyze → Evaluate → Create).
I need deep research on:
1. HOW BLOOM'S TAXONOMY IS IMPLEMENTED IN SOFTWARE:
- What does a Bloom's-tagged activity/lesson look like in a database schema?
- How do existing platforms (Toddle, IXL, Khan Academy, Synthesis Tutor) map activities to Bloom's levels?
- Is there a standard taxonomy ontology or tagging system used across EdTech?
- How does adaptive learning use Bloom's levels to decide what to show next?
2. CHILD-LED LEARNING UX PATTERNS:
- What does "the student chooses" actually look like as a user interface for an 8-year-old vs a 14-year-old?
- How do Montessori digital platforms handle self-directed learning?
- How do Reggio Emilia-inspired platforms document the child's learning journey?
- What's the UX difference between "choose your adventure" (branching) vs "open world" (free exploration) vs "guided inquiry" (Socratic)?
- How do you prevent choice paralysis in young learners?
- Real examples from: Synthesis Tutor, Gooru, KhanMigo, MindJoy, Toddle, any Montessori LMS
3. PROGRESS TRACKING WITHOUT TRADITIONAL GRADES:
- Competency-based progression vs age-based progression
- How do you visualize Bloom's level achievement to a student, a parent, and a teacher?
- Portfolio-based assessment: what works, what's just busywork?
- How do UK councils assess AP/SEND progress for continued funding? What data format do they need?
- EHCPs (Education, Health and Care Plans) — what does the software need to generate?
4. THE "SCREENING" INTERFACE (Ages 4-8):
- How do you screen for neuroquirks digitally without it feeling clinical?
- What validated screening tools exist that could be embedded in a platform?
- Strengths and Difficulties Questionnaire (SDQ) — can it be gamified?
- How does the Finnish early years model handle developmental assessment?
Cite all sources. Include links to specific products, papers, or open-source repos where these patterns are implemented.
PROMPT 3: Alpha School Deconstruction + Lessons Learned for GenEvolve
Alpha School (Austin, Texas, founded by MacKenzie Price and Elon Musk-adjacent figures) is an AI-first alternative school that became controversial due to parent backlash. I need a comprehensive deconstruction:
1. TECHNOLOGY STACK:
- What is "Incept"? Is it proprietary or built on existing platforms?
- What is "Timeback"? How does it measure time saved vs traditional learning?
- Which external platforms do they wrap? (IXL, Khan Academy, Amplify, Newsela — verify each)
- How do they sequence AI tutoring vs human coaching?
- What's their actual daily schedule? (2 hours AI learning + rest = projects/sports?)
2. THE PARENT REVOLT:
- What specifically went wrong? Compile all public complaints (Reddit, X/Twitter, news articles, parent forums)
- Was it the tech, the pedagogy, the communication, or the price?
- What did parents expect vs what they got?
- How did Alpha respond?
- What's their current enrollment/retention status?
3. WHAT THEY GOT RIGHT:
- Time compression (2hr core learning claim) — is there evidence this works?
- Coaching model instead of teaching — what's the research?
- Entrepreneurship integration — what does this actually look like in practice?
- Their marketing and brand (they got massive attention — how?)
4. LESSONS FOR GENEVOLVE:
- GenEvolve is building a similar model in the UK: 2hr morning academics + afternoon projects/real-world
- But GenEvolve emphasizes: physical-first (no screens ages 4-8), child-led (not AI-led), Bloom's taxonomy, wellbeing/SEND, community village model
- What specific mistakes should GenEvolve avoid based on Alpha's experience?
- What should GenEvolve copy?
- How should GenEvolve differentiate in messaging to avoid the same backlash?
5. OTHER AI-FIRST SCHOOLS:
- Synthesis (founded by ex-SpaceX education lead) — how is it different from Alpha?
- Primer (Silicon Valley micro-school network)
- Sora (virtual high school, project-based)
- Any UK equivalents attempting similar models?
Cite all sources with links. Include Reddit threads, news articles, X/Twitter posts, parent reviews.
PROMPT 4: UK Alternative Provision (AP) & SEND — Platform Requirements for Council Funding
I'm building an education platform that will serve Alternative Provision (AP) and SEND students in the UK, alongside neurotypical students. The platform needs to generate the data and reports that local authorities require to justify continued funding (up to £140,000/child/year for complex out-of-area residential placements).
Deep research needed:
1. WHAT LOCAL AUTHORITIES NEED TO SEE:
- What progress data do councils require for AP placements? (weekly, termly, annually?)
- What format do EHCP (Education, Health and Care Plan) annual reviews require?
- What metrics prove "the intervention is working" for AP/SEND?
- How do councils measure "distance travelled" for students who are below expected levels?
- What does Ofsted look for when inspecting AP/alternative provision? (their specific quality indicators)
- What data format does the DfE's "Apply to be on the register of independent schools" require?
2. EXISTING SOFTWARE FOR AP/SEND REPORTING:
- What platforms do UK AP schools currently use? (SIMS, Arbor, Bromcom, Provision Map, EHCPlan, Edukey?)
- What are their limitations?
- Is there an API or data standard for EHCP data exchange between schools and councils?
- What does EDClass (DfE-accredited alternative provision) use for their reporting?
3. SEND SCREENING AND ASSESSMENT:
- What validated screening tools are used in UK schools for neurodiversity? (SDQ, Boxall Profile, PASS, CAT4?)
- Can these be embedded in a platform or do they require licensed administration?
- How do you screen ages 4-8 without diagnosis (early identification vs pathologization)?
- What's the UK legal position on schools doing their own screening vs requiring educational psychologist assessment?
4. FUNDING MECHANICS:
- How does the High Needs Block work? (DSG allocation, top-up funding, place funding)
- What's the process for an AP school to receive council funding?
- Can a new school (like GenEvolve's Education Village) access AP funding before achieving Ofsted rating?
- How does the new HNPCA (High Needs Provision Capital Allocation) grant work? (Surrey just got £16.14M)
- What's the timeline from opening to receiving first council-funded placements?
5. PLATFORM REQUIREMENTS MATRIX:
Based on all the above, produce a requirements specification for what the platform must track, generate, and report. Include:
- Student profile data fields
- Progress tracking frequency and format
- EHCP integration requirements
- Attendance and engagement metrics
- Wellbeing/behavior tracking (positive, not punitive)
- Parent portal requirements for AP/SEND
- Data export formats (XML, CSV, API endpoints?)
- GDPR and ICO Children's Code compliance requirements
Cite all sources — Ofsted guidance, DfE statutory guidance, council published requirements.
PROMPT 5: Education Village Platform — Physical-Digital Bridge & Parent Portal Design
I'm designing the technology platform for an "Education Village" — a physical school site (180 acres, ages 4-18) that also has hybrid and fully online tiers. The platform is the "connective tissue" between the physical experience and the digital record. The school follows a Finnish-inspired model: ages 4-8 are minimal/no screen; ages 8-12 begin digital literacy; ages 12-18 use the platform fully.
Research needed:
1. PHYSICAL-DIGITAL BRIDGE:
- How do outdoor education programs log activities digitally? (Forest School, Duke of Edinburgh, Scouts digital badges)
- What IoT or low-tech solutions exist for recording physical learning without giving young children screens? (teacher check-ins, NFC tags, QR code stations, photography, voice recording?)
- How do Montessori schools document "works" (activities) digitally?
- How does Geelong Grammar's Timbertop (zero-tech Year 9 campus) document student progress?
- What does a "digital twin" of a physical learning space look like? (Any examples from education villages, innovation hubs, makerspaces?)
2. PARENT PORTAL — WHAT WORKS:
- What do parents of AP/SEND children specifically need to see? (vs neurotypical parents?)
- How do you show "my child is thriving" without traditional grades?
- Best-in-class parent portals: ClassDojo (what they got right), Toddle (portfolio approach), Seesaw (activity feed), Tapestry (early years), Evidence Me
- What makes parents trust a school platform vs feel surveilled?
- How do you handle separated/divorced parents with different access rights?
- Real-time vs weekly digest — what do parents actually prefer?
- How do you communicate wellbeing/emotional development without violating the child's privacy?
3. THE THREE-TIER ARCHITECTURE:
- Tier 1: Physical on-site students (platform = record-keeping, portfolios, parent comms)
- Tier 2: Hybrid students (home-ed + 2-3 days on-site, platform = learning delivery + community)
- Tier 3: Fully online students (platform = complete learning environment)
- How do you design one platform that serves all three without the on-site experience feeling "lesser" than the online one?
- What features should be tier-specific vs universal?
- How do hybrid models handle the "two days at home" content? (Asynchronous? Live? Project-based?)
4. COMMUNITY FEATURES:
- GenEvolve's model includes: peer mentoring, inter-age collaboration, parent education, industry partnerships
- What community platforms work in education? (Circle, Mighty Networks, Discourse, custom?)
- How do you build community for families, not just students?
- Safeguarding requirements for under-18 community features (DfE Keeping Children Safe in Education)
- How do you enable student collaboration without creating a social media dynamic?
5. MINIMUM VIABLE PRODUCT SCOPE:
Based on all research, define the MVP feature set that could launch in 4-6 weeks with AI coding agents, covering:
- Ages 8-12 digital onboarding (not 4-8, they're screen-free)
- Basic portfolio/progress tracking
- Parent portal (view-only initially)
- Teacher/coach dashboard
- Bloom's taxonomy activity tagging
- SEND/AP reporting (council-ready)
- Single sign-on (no Google/Microsoft — use passkeys or email magic links)
- Mobile-first responsive design
Produce a prioritized feature list with "Must Have" / "Should Have" / "Nice to Have" categories.
Cite sources, link to specific products and implementations, include screenshots or descriptions of best-in-class UX patterns.
NEW PROMPTS (v2 — Feb 28, 2026)
PROMPT 6: Shelley's Competitive Landscape — Deep Dive on All Platforms
I'm advising the founder of GenEvolve, a UK "Education Village" initiative building a physical + hybrid + online school for ages 4-18. The founder has shared a list of platforms and schools she's been researching. I need a comprehensive competitive analysis of each one.
For EACH platform/school below, provide:
- What it is, who founded/runs it, when founded
- Business model and pricing (tuition, SaaS subscription, freemium?)
- Student numbers and geographic reach
- Technology stack if discoverable
- Pedagogy/philosophy (traditional curriculum? alternative? child-led?)
- Key differentiator vs. traditional schools
- Strengths and weaknesses
- Data sovereignty position (self-hosted? cloud? where is student data?)
- SEND/neurodiversity support
- How it positions against GenEvolve's unique model (physical village + child-led + Bloom's + sovereign data + community "We Economy")
PLATFORMS/SCHOOLS TO ANALYZE:
1. Almach Lite (almach-lite.com) — AI-powered teaching/learning resource from UAE (DIFC registered), currently in beta. NOT a curriculum-hosting platform. Designed for educators to create content. Website: almach.ai and almach-lite.com
2. Toddle (toddleapp.com) — Full-stack school platform used by 2000+ schools. Some of GenEvolve's team use it but prefer MindJoy. WHY does MindJoy win over Toddle for this context?
3. MindJoy (mindjoy.com) — AI STEM education, ages 8-15, founded by teachers during COVID. GenEvolve has a demo booked for March.
4. King's InterHigh (interhigh.co.uk) — UK's largest online school, part of Inspired Education Group. Full British curriculum, primary through A-Level, 3000+ students.
5. Inspired Education Group — The parent company of King's InterHigh. Global premium school network. How big? How many schools? What's their tech/platform strategy?
6. Highgrove Education / Highgrove Online School — UK online school. Research their model, pricing, curriculum approach.
7. Eton College — What is Eton doing in online/digital education? (EtonX was their online platform — is it still active? What replaced it?)
8. Oxford Online School — UK online school. One family considered this during COVID for home education.
9. Sophia High School (sophiahigh.school) — "A World Class British Online School." Founded by Melissa McBride. Ages 4-16. Metaverse/immersive learning angle.
10. School of Humanity (schoolofhumanity.org) — International online school with hybrid hubs in Dubai and Florida. "Similar ethos" to GenEvolve but "not our 360 village concept." DEEP DIVE — this is the closest philosophical competitor.
11. Nord Anglia Education — 80+ schools globally, "luxe" international education brand. Professional operation with a "village school" model. Has an entrepreneurial element. How big? Revenue? Tech platform?
12. GEMS Education (Sunny Varkey) — "Much much larger than Nord Anglia" and "second place for international luxe education." Founded by Sunny Varkey. How many schools? Revenue? Reputation?
13. Guildford High School (part of GDST — Girls' Day School Trust) — They are reporting that AI integration is "excellent" BUT girls don't trust it. There have been errors in the platform's responses. Girls "prefer to talk to a human or a bot." This is a data point about AI adoption in education — research what GDST is doing with AI across their schools.
KEY FRAMING from GenEvolve's founder:
"All these high end education groups - who all offer UK curriculum and a luxe desirable brand - all drive their students based on ego and competing for admissions against each other - driving league tables and the typical 13 * GCSEs attainment."
GenEvolve's differentiation: community over competition, wellbeing over grades, child-led over prescribed, physical village over online-only, data sovereign over Big Tech dependent.
Produce a competitive positioning matrix and threat assessment (High/Medium/Low/None) for each relative to GenEvolve. Identify which are true competitors, which are potential partners/tools, and which are "what GenEvolve is NOT."
PROMPT 7: Build vs Buy in 2026 — The AI Development Thesis Applied to Education Platforms
Context: I'm advising an education startup (GenEvolve, UK) on whether to build, buy, or fork their learning platform. The traditional advice is "adopt open source like Moodle, save £500K+ vs building from scratch." But the world has changed:
KEY EVIDENCE:
- Theo Browne (t3.gg, ex-principal engineer at Twitch) published "Software engineering is dead now" (Feb 2026). His coding agents cloned the functionality of Frame.io (a video review/approval app that Adobe acquired for $1.2B) in two weeks while he was working on other projects.
- The marginal cost of software development is heading toward zero.
- AI coding agents (Claude Code, Codex, Cursor, Windsurf) can scaffold entire applications in days/weeks.
- Infrastructure costs have collapsed (Vercel, Fly.io, Railway, Hetzner — pennies per user).
RESEARCH NEEDED:
1. THE "BUILD OVER BUY" THESIS:
- What evidence exists (beyond Theo) that AI-assisted development has fundamentally changed build vs buy economics?
- What are real examples of companies/individuals building custom tools that replace expensive SaaS?
- How does this apply specifically to education/LMS platforms?
- What's the realistic timeline for building a custom education platform with AI agents in 2026?
2. OPEN SOURCE LICENSE REALITY CHECK:
- AGPL (Canvas, Open edX) — what EXACTLY are the commercial restrictions? Can you fork and sell as SaaS?
- GPL (Moodle, Chamilo) — the "SaaS loophole" — does it actually work for a school platform?
- MIT (LearnHouse) — truly permissive, but is the codebase mature enough?
- Apache 2.0 (OpenOlat) — permissive but wrong tech stack (Java)?
- If you plan to white-label and license the platform globally, which licenses allow this WITHOUT releasing your source code?
3. THE SECURITY ARGUMENT:
- Open source platforms (Moodle especially) have hundreds of known CVEs. They're high-value targets.
- A custom-built, minimal-surface-area platform is theoretically less attackable but has no battle-tested security community.
- For CHILDREN'S DATA in the UK (ICO Children's Code, GDPR), which approach is actually safer?
- What are the real-world breach statistics for education platforms? (Moodle breaches, Canvas breaches, custom school platform breaches?)
- How does self-hosting change the security calculus vs. SaaS?
- Reference: Equifax (147.9M Americans, 15.2M UK residents exposed), Google/Facebook/Apple 2025 credential breach (16B credentials)
4. THE DATA SOVEREIGNTY ARGUMENT:
- UK Children's Code requirements for education platforms
- Why parents choosing alternative education are ALREADY privacy-conscious
- Self-hosting as a marketing differentiator (trust = competitive advantage)
- What does "sovereign by design" look like architecturally?
- How do you prove data sovereignty to a council (Surrey CC) that's evaluating your school?
5. RECOMMENDATION FRAMEWORK:
Given all the above, produce a decision framework for GenEvolve:
- When does "fork and customize" beat "build from scratch"?
- When does "build from scratch with AI agents" beat "fork"?
- What's the minimum viable team size in 2026? (Thesis: 1-2 devs + AI, not a traditional team)
- What's the realistic cost comparison? (Fork Moodle vs Fork LearnHouse vs Build custom vs Buy SaaS)
- What's the timeline comparison?
- What should the CTO's first 90 days look like?
Include real numbers, real timelines, real examples. This is for a technical audience (the CTO candidate) not a pitch deck.
PROMPT 8: Finnish Early-Years Model — Deep Dive for Education Village Design
GenEvolve is building a UK "Education Village" for ages 4-18. The early years (ages 4-8) are explicitly modeled on Finnish pedagogy: play-based, minimal/no screens, nature-integrated, child-led. I need comprehensive research on how Finland actually does this and how it maps to a UK context.
1. FINNISH EARLY YEARS IN DETAIL:
- What does a typical day look like for a Finnish 5-year-old? Hour by hour.
- How much outdoor time? What activities? In what weather conditions?
- When does formal reading/writing instruction begin? (Age 7 in Finland — how does this work in practice?)
- What's the teacher-to-child ratio?
- How are Finnish early years teachers trained? (Master's degree requirement — what does the curriculum cover?)
- What assessment methods are used? (No standardized testing — so what instead?)
- How is "readiness" determined without testing?
2. PLAY-BASED LEARNING EVIDENCE:
- What does the research say about play-based vs. formal instruction for ages 4-8?
- Cambridge Primary Review, EPPE study, Scandinavian longitudinal studies — what do they show?
- Does delaying formal academics (Finnish model) produce better or worse outcomes by age 12? By age 18?
- What's the evidence on screen time for under-8s? (WHO guidelines, UK CMO guidance, Finnish policy)
3. NATURE-BASED EDUCATION:
- Forest School model (UK version of Scandinavian outdoor education) — what does it actually look like?
- How do Finnish "nature kindergartens" operate?
- Evidence on nature exposure and child development (attention, wellbeing, creativity)
- How do you run outdoor education in British weather year-round?
- Insurance, safeguarding, and risk assessment requirements for outdoor learning in UK
4. MAPPING TO UK CONTEXT:
- EYFS (Early Years Foundation Stage) — how does Finnish model align with UK statutory requirements?
- Can a UK school delay formal reading instruction to age 7 and still meet EYFS goals?
- What UK schools already use Finnish-inspired approaches? (Any case studies?)
- How does Ofsted view play-based/outdoor provision in the early years?
5. TECHNOLOGY FOR NON-SCREEN EARLY YEARS:
- How do you record and track child development without giving children screens?
- Teacher observation tools (paper-based → digitized by teacher, not child)
- Photography/video as documentation (Reggio Emilia "hundred languages" approach)
- Voice notes and audio recordings
- How does this data feed into the platform for parent visibility and council reporting?
6. PRACTICAL DESIGN FOR GENEVOLVE:
- What physical spaces does a Finnish-inspired early years setting need?
- Indoor: how should classrooms be designed? (Montessori prepared environment? Reggio atelier?)
- Outdoor: what natural features, equipment, shelter?
- Staffing model: teacher + teaching assistant ratios for ages 4-8 with SEND inclusion
- How does the "village" aspect (intergenerational, community) enhance early years?
- Specific daily schedule proposal for GenEvolve ages 4-8
Cite sources including Finnish National Agency for Education (OPH), UK EYFS framework, research papers, and case studies from UK schools using Finnish-inspired models.
USAGE NOTES
Original prompts (1-5): Core platform architecture and design research - Prompt 1 → Technical platform comparison (already mostly done — use for validation) - Prompt 2 → Pedagogy-to-software bridge (hardest design problem) - Prompt 3 → Alpha School playbook (what to steal, what to avoid) - Prompt 4 → Business case (council funding requirements) - Prompt 5 → Product design (physical-digital bridge + parent portal)
New prompts (6-8): Competitive intelligence and strategic positioning - Prompt 6 → Shelley's competitive landscape (all 13 platforms/schools) - Prompt 7 → Build vs Buy thesis with Theo's AI development evidence - Prompt 8 → Finnish model deep dive for village design
Recommended execution order: 1. Prompt 6 first (competitive landscape — Shelley is waiting for your take) 2. Prompt 7 next (build vs buy — the strategic argument you'll present at coffee) 3. Prompt 8 (Finnish model — foundational for the village design conversation) 4. Then 2 → 4 → 5 for product specification depth
Recommended tools: - Gemini Deep Research → Prompts 1, 4, 5, 6, 8 (factual, regulatory, comparative) - ChatGPT Deep Research → Prompts 2, 3, 7 (creative, UX patterns, social sentiment, thesis-driven) - Perplexity Pro → Quick follow-ups on any gaps
Tip: Feed the output of Prompt 6 into Prompt 7 for best results — understanding the competitive landscape informs the build vs buy decision.