🎓 EduVerse: AI-Powered Personalized Education Platform

[amardeep]

Live Link: https://www.eduverse.network/

Problem Statement

Standardized education overlooks individual learning styles and paces, leading to disengagement and hindering student potential. Students and educators face the challenge of a rigid system that doesn’t adapt to diverse needs, impacting learning outcomes.

---

Solution Overview

Our AI-Powered Personalized Education Platform offers a dynamic solution by employing a multi-agent system built on the Alith Agentic Framework. Specialized AI agents collaboratively analyze individual learning styles, curate tailored content, provide adaptive tutoring, and track progress in real time. This creates a uniquely personalized learning journey that adjusts to each student’s needs and pace. By integrating diverse educational resources, the platform aims to enhance engagement, improve learning outcomes, and empower both students and educators.

---

Project Description

This platform is designed to benefit a wide range of users across the educational landscape. Students of all ages, from kindergarten through higher education, can experience a more engaging and effective learning journey tailored to their specific needs. Educators can utilize the platform to enhance their teaching capabilities, gain deeper insights into student progress, and free up time for more individualized support.

On-Chain Components: A First-Class Implementation and Core Requirement for Blockchain Deployment

Our AI-powered personalized education platform is not merely “blockchain-enhanced” — its core functionality and unique value proposition depend fundamentally on blockchain deployment via Hyperion. The decentralized, trustless, and immutable nature of blockchain—particularly Hyperion’s AI-optimized environment—is essential to realizing our vision.

---

On-Chain Achievement Verification (Smart Contract Architecture)

What’s On-Chain:
The LearningRecord.sol smart contract serves as an immutable, on-chain ledger for all student accomplishments.

• Achievement Mapping: The contract maintains a mapping from a user’s wallet address to an array of Achievement structs, where each struct contains the moduleName and a timestamp.
• Secure, Off-Chain Verification: A secure, server-controlled wallet is the only address authorized to call the addAchievementWithSignature function. This ensures achievements are only recorded after the platform’s backend has verified the user’s quiz completion and signature.

Why It Needs Blockchain:

• Trustless & Immutable Credentials: By recording achievements on the Metis Hyperion testnet, we create a permanent, tamper-proof record of learning that is owned by the user and verifiable by anyone (e.g., employers, other institutions) without relying on a centralized database.
• Data Sovereignty: Students retain full control of their learning data. On-chain metadata ensures ownership and transparency, unlike siloed centralized platforms.
• Foundation for a Trustless Ecosystem: This on-chain record is the foundational layer. While AI agents currently operate off-chain for performance, their most critical output—the certification of learning—is secured on the blockchain.

AI Enablement via the Alith Agentic Framework

What is Alith:
Alith is a modular, multi-agent framework designed to power personalized learning by simulating human-like reasoning, memory, and collaboration among AI agents. It forms the intelligence layer of our platform, with deep interoperability with blockchain systems like Hyperion.

Key Features of Alith (Implemented):

• Agent Management: Modular deployment of specialized AI agents:
  • Learning Style Analyst (Implicit in course selection)
  • Content Curator (AI Study Guide Generator)
  • Personalized Tutor (Quiz Hints & Tutor Chat)
  • Progress Tracker (On-chain achievement logging)
• Persistent Memory: Agents maintain long-term memory and context across sessions (demonstrated in Telegram bot), enhancing personalization.
• Toolchain Access: Seamless access to external APIs and knowledge bases for generating content and hints.

Blockchain Integration with Alith:
The Alith agent framework operates off-chain to provide a responsive and intelligent user experience. The blockchain is used as the ultimate source of truth for the results of these AI interactions.

• Verifiable Outcomes: When a user successfully passes a quiz, the off-chain backend coordinates with the ProgressTrackerAgent to commit this achievement to the LearningRecord smart contract, creating an immutable record✅.
• Foundation for On-Chain Logic: The current architecture provides the groundwork for future enhancements where agent workflows could be triggered and validated by smart contracts.

Secure, Real-Time AI Inference (Leveraging Hyperion’s AI-Native Infrastructure)

What’s On-Chain (or Hyperion-enabled):
While heavy AI workloads (e.g., model training, long-form inference) run off-chain via the Alith framework, Hyperion enables:

• Verifiable AI Outputs: AI-driven assessments and tutor recommendations are validated off-chain, with the final “proof-of-completion” immutably stored on-chain. This ensures educational integrity and transparency—unlike opaque, centralized AI systems.
• Low Latency, High Throughput: Hyperion’s parallel execution environment supports the fast transaction finality needed to record achievements in near real-time, keeping learners engaged.

Data Sovereignty, Incentives, and Trust in Our Education Platform

Our AI-Powered Personalized Education Platform leverages Hyperion’s blockchain to give students true control over their learning data and foster a trusted educational environment.

Wallet-Based Identity and User Control
In our platform, a user’s Web3 wallet (e.g., MetaMask) serves as their identity. Authentication is handled securely via message signing, ensuring users control access to their accounts without traditional passwords. This wallet-centric approach is the first step toward a future of true data sovereignty, where learning profiles can be fully controlled by the user.

Importance of Trust and Transparency
In education, trust and transparency are paramount:

• Learner Trust: Students need to trust that their achievements are real. On-chain verifiable credentials build this crucial trust.
• Empowering Educators: Transparent data provides educators with reliable insights into student progress and curriculum effectiveness.
• Equity & Accessibility: Blockchain prevents fraud and offers a universal, trusted way for students to showcase skills, democratizing opportunities regardless of background.

Community Engagement Features

• Peer Study Groups: Connect with others for collaborative discussions and support.
• Q&A Forums: Ask and answer subject-related questions to deepen understanding.
• Simulated Collaboration: Practice teamwork in virtual, project-based learning scenarios.
• Optional Mentorship: Receive guidance from experienced peers and educators.
• Progress Sharing (Privacy-Controlled): Share milestones to motivate peers while maintaining control over visibility.
• Community Challenges: Participate in gamified group activities to promote engagement and friendly competition.

---

Future Roadmap & Planned Enhancements

This MVP is the foundation for a much larger vision. Our planned enhancements include:

1. Token-Based Incentives ($LP)

   • Introduce a utility token, LearnPoints ($LP), to create a Learn-to-Earn (L2E) model that rewards students for completing modules, mastering skills, and contributing to the community.
   • $LP could be staked for premium features or used to vote on platform decisions, aligning community interests.

2. Decentralized Agent Marketplace

   • Allow community developers to contribute new agents (e.g., specialized tutors or regional content curators).
   • Agents will be published and authenticated on-chain with verifiable capabilities.
   • Token-based reputation and staking mechanisms will prevent misuse or malicious behavior.

3. On-Chain Agent Coordination & Composable Workflows

   • Evolve our smart contract architecture to manage agent workflows directly, enabling fully trustless and transparent educational logic.
   • Enable educators to compose reusable learning workflows by chaining agents together via simple declarative schemas, backed by on-chain validation.

4. ZK-Verifiable Inference

   • Integrate zkML (zero-knowledge machine learning) to enable cryptographic verification of off-chain AI outputs. This allows validators and institutions to confirm an agent’s decision (e.g., a test score) without exposing the underlying model or input.

5. Multilingual & Accessibility Agents

   • Create plug-and-play AI agents specialized in localization, translation, and neurodiverse learning strategies.

Team Members

• @amardeep
• @priyankg3

GitHub

https://github.com/amardeepio/Eduverse

Video tutorial EduVerse tutorial video

Presentation and future roadmap: Presentation and RoadMap

[arbilearn]

This is surely a great concept, waiting to see it come to live

[ThalaZikol]

Impressive project! The combination of multi-agent AI and Hyperion blockchain creates a personalized, transparent, and trustworthy education platform. Features like SSI, Learn-to-Earn, and zkML show long-term vision and innovation. Looking forward to seeing EduVerse grow!

[Keir]

bu özel bilgiler ıcın teşekkürlerrr

[ravisharma]

Really like the detailed proposal..

Hope your project select for build.

Best wishes.

[Shruti]

education is a very important part of our life, glad you touch the real problem and trying to solve this with the help of AI

[arbilearn]

Great idea, Happy Hacking

[deadman_xbt]

This is by far the best proposal of hyperhack , right now.

impressed by the usage of alith and AI in education category.

hope it turned out to be a super product and solve the problems.

good luck

[heromatters]

Hello @amardeep ,

First of all, thank you for presenting the proposal on such an important narrative — education.

I have few questions :-

1. How does EduVerse plan to balance AI personalization with curriculum standardization, especially for formal education systems that require fixed syllabi and assessments?
2. How does EduVerse ensure a smooth experience for users who are new to both AI and blockchain (students, teachers, or even schools)?

[amardeep]

@heromatters

Hello @amardeep ,

First of all, thank you for presenting the proposal on such an important narrative — education.

Thank you so much for your kind words and for taking the time to go through the proposal.

I would like to answer your both questions

Balancing Personalization & Standardization:

EduVerse’s AI works within the set curriculum, not outside it. It helps students learn at their own pace and style while making sure they cover everything required. For teachers, it highlights exactly where students need help, so they can focus their efforts more effectively—without losing control of the classroom.

Helping New Users Feel at Home:

We keep things simple. Instead of talking tech, we show the benefits—like personalized learning and instant feedback. The platform is easy to use, with gamified learning for students and clear dashboards for teachers. Blockchain runs quietly in the background—users just see that their records are secure and portable. Plus, we offer helpful guides and real human support.

Feel free to ask further.

Thank you.

[Cryptopotato]

Hello @amardeep

Why did you choose Hyperion specifically for EduVerse instead of other chains like NEAR or Avalanche that also offer AI integrations and scalability?

What strategies are used to detect and correct AI biases in agent decision-making (e.g., tutoring paths, performance assessments)?

[amardeep]

Thanks for your questions.

Why Hyperion for EduVerse?

We chose Hyperion over other chains like NEAR or Avalanche because it’s built for AI-native Web3 applications, perfectly fitting EduVerse’s needs:

• Native AI Optimization: Hyperion’s MetisVM and AI co-processor are designed to efficiently run and verify AI workloads directly on-chain. This means more trustworthy and cost-effective AI decisions compared to other chains that simply “support” AI.
• Direct Alith Integration: Hyperion has a direct, announced integration with the Alith Agentic Framework, which is the core of our multi-agent system. This makes developing and deploying our AI agents much smoother.
• Real-Time Performance: Its parallel execution and high throughput (with MetisDB for data) ensure a super responsive and scalable learning experience, crucial for instant AI feedback and interactions.
• Cost-Efficiency: As a Layer 2, Hyperion offers lower transaction costs, making expensive AI operations more viable for a large educational platform.

How We Detect & Correct AI Biases?

• Diverse Training Data: We’ll use very diverse datasets to train our AI, making sure it learns from a wide range of students and backgrounds. We’ll also use techniques like data balancing to prevent underrepresentation.
• Fairness-Aware Algorithms: We’ll build our AI models using algorithms specifically designed to reduce performance disparities across different groups, constantly monitoring with fairness metrics.
• Human Oversight: Educators and experts will review AI-suggested paths and assessments. Their feedback will directly help refine the AI.

Thank you

[amardeep]

Project update:

Github : GitHub - amardeepio/Eduverse

project update:

Hey everyone,

When you click that “Complete Module” button, something really cool happens behind the scenes. Think of it like getting a digital diploma notarized on the blockchain.

Here’s the simple 3-step process:

1. You Sign the Diploma: Your MetaMask wallet pops up and asks you to “sign” a message. This is like putting your unique, unforgeable signature on the diploma, proving you completed the work. It’s completely free and doesn’t cost any gas.
2. A Secure Courier Delivers It: Our server acts as a secure courier. It takes your signed diploma and delivers it to the blockchain to be officially recorded.
3. The Blockchain Notarizes It: The smart contract acts as a public notary. It mathematically verifies that your signature is authentic. Once verified, it permanently stamps your achievement into its public ledger.

That’s it! The “Achievement Unlocked” message means your accomplishment is now a permanent, verifiable part of your learning history that you, and only you, control with your wallet.

Thanks

@pavel @t0mcr8se @daryl

---

[daryl]

Great update @amardeep !

[CrisMetis]

Really cool idea! I believe putting together multi-agent AI with the Hyperion blockchain makes EduVerse feel both smart and trustworthy. Love the direction with features like Learn-to-Earn. Really exited to see where it goes!

[amardeep]

Thanks a lot for the appreciation.

[sumankr]

Very Nice Idea! @amardeep

[amardeep]

EduVerse Progress Update
Hey everyone,

We’ve made progress on the EduVerse platform, and we’re excited to share what’s now fully functional! We’ve built and integrated the core components of our vision for a personalized, on-chain educational experience.

---

Full User Login & Session Management
Users can securely Sign-In with Ethereum via MetaMask.
Creates a verifiable session to track progress — no email/password required!

---

Dynamic Course & Quiz System
Clean, tile-based layout for all available courses.
Users can browse topics, pick a course, and take a multi-question quiz.

---

On-Chain Achievements
Upon quiz completion, users sign a message via wallet.
The ProgressTrackerAgent sends this to our smart contract.
A verifiable credential (your on-chain diploma) is minted on the Hyperion Testnet.
Users receive a transaction link to view their credential instantly!

---

Multi-Agent AI Backend (Alith Framework)
We’ve evolved into a true multi-agent system, including:

API Gateway – Handles all user requests.
Agent Service – Coordinates our AI agents.
TutorChatAgent – Powered by Alith + DeepSeek + Gemini, provides smart, conversational, context-aware help for any course.
QuizHintAgent – Offers conceptual hints during quizzes (no spoilers!).

---

Enhanced User Experience
Added typewriter animation to AI responses for a more human-like feel.
Loading animations enhance responsiveness and polish.

---

We’re proud of how far we’ve come — but this is just the beginning. EduVerse is becoming a living, breathing learning companion for the decentralized world!

Let us know your thoughts, feedback, or ideas!

Screen capture of progress till now

Thank you

[amardeep]

Screen capture video link in good quality Eduverse screen capture

[heromatters]

This is truly amazing amar and thanks for sharing the update.

The UI looks so cool.