Brain Ink

Demo

About

The problem it solves 1. Lack of Trust, Ownership & Monetization in EdTech Current educational platforms centralize content and profits, leaving creators without fair ownership and students with no transparent reward systems. Creators can’t verify ownership of content, and often don’t earn proportionally to their impact. Learners have no on-chain proof of participation, credentials, or contributions. There's no universal token economy to reward student progress, tournament wins, or knowledge-sharing efforts. Existing platforms limit the transferability of digital credentials or assets. BrainInk solves this by using the blockchain (Base chain) to tokenize achievements, learning activity, and content ownership — enabling direct monetization, verification, and transferability of educational assets. 2. Fragmented, Isolated Messaging in Learning Platforms Most platforms treat messaging as an afterthought, separating it from actual learning. There's no integration between chats and study content, making collaboration inefficient. Students jump between tools like WhatsApp, Discord, and Google Docs — which breaks focus and increases distractions. Educators can’t embed AI or learning support directly within communication tools. BrainInk integrates real-time messaging directly with AI (K.A.N.A.), group study, and tournament updates — creating a central hub for all academic and social interactions. 3. Inequitable Access to Quality Tools in Africa & Emerging Markets High-quality AI tutors, verified educational libraries, and personalized learning systems are often locked behind expensive paywalls. There’s no platform combining learning, social engagement, creator monetization, and blockchain rewards — all in one. BrainInk democratizes access with a $5 monthly subscription in Africa, and $10 globally, giving every learner and teacher access to premium tools and transparent earning models. In Summary BrainInk reimagines learning for the digital-native generation — decentralized, social, gamified, and fair. It solves the problem of: Unrewarded learning Unverifiable digital achievements Unmonetized content creation Scattered communication in academic spaces All by building an onchain educational ecosystem with AI-powered collaboration at the center.

How it's made

Challenges we ran into Real-time AI + Messaging Integration Challenge: We wanted users to seamlessly chat with our AI assistant (K.A.N.A.) within the same interface as their friend and group conversations. Integrating real-time messaging with an always-on AI, while ensuring low latency, context memory, and multimodal file support, proved complex. How We Solved It: We decoupled AI message parsing from standard chat WebSocket streams. Introduced a queuing system and fallback responses when latency spiked. Used streaming token responses (à la OpenAI's chat streaming) to keep it feeling real-time. Created isolated AI agents per conversation thread with scoped memory. 2. 🔐 Blockchain Wallet Integration for Non-Crypto Users Challenge: Many users, especially students, were unfamiliar with Web3 tools like MetaMask or the idea of tokens. This caused friction during onboarding and transaction testing. How We Solved It: We implemented an embedded wallet system using a social login layer (Firebase + third-party custodial service) for users who don't want full wallet control. For advanced users, we added optional MetaMask/WalletConnect flows. Built a test token faucet and sandbox mode so users could learn the system risk-free. 3. 📚 Indexing and Searching 200M+ Academic Papers via CORE API Challenge: We integrated the CORE API for academic paper searches, but hit rate limits and slow query returns when testing at scale. How We Solved It: Cached search results by subject and region to reduce duplicate requests. Preloaded high-frequency academic topics. Built a local proxy server that handles retries and exponential backoff. 4. 🧩 Cross-Feature Dependencies in Development Challenge: With a small team of two developers (each working with an AI assistant), features like quizzes, tournaments, and K.A.N.A. often depended on shared modules like XP systems or leaderboard logic. This created merge conflicts and delays. How We Solved It: Created strict module boundaries using a mono-repo structure. Used feature flags to allow partial deployment of dependent systems. Wrote shared logic as external libraries consumed by both teams, ensuring autonomy but compatibility.