🎣 HookED – Learning That Slaps
A gamified microlearning app for neurodivergent minds and curiosity-driven learners—powered by the Perplexity Sonar API.
In One Line
HookED turns learning into a real-time, scrollable experience—powered by AI, structured by story, and gamified for curiosity.
🌟 Inspiration
"What if learning felt like scrolling TikTok—but actually made you smarter?"
HookED was born from frustration with traditional learning and a desire to create something that speaks to how people actually absorb information—especially those with ADHD, burnout, or short attention spans.
We saw the power of stories, analogies, and visual reinforcement—and paired that with cutting-edge AI to build a new way to learn.
🚀 What It Does
HookED is a mobile-first app that delivers:
- 📌 60-word hooks powered by the Perplexity Sonar API
 - 📚 Real-time sources and citations
 - 🧠 Instant quizzes to reinforce key ideas
 - 🧭 Optional “Hook Quests” for deeper exploration
 
Along the way, users collect XP, badges, and streaks—making learning feel like leveling up in a game.
🏗️ Architecture Overview
| Component | Technology Used | Purpose | 
|---|---|---|
| Mobile App | React Native | Custom UI for onboarding, search, quizzes, Hook Feed, and quests | 
| Backend | Python + FastAPI | Routing, XP logic, profile management, and API calls | 
| Database | MongoDB Atlas | Stores users, history, XP, quest states, and badge data | 
| AI Engine | Perplexity Sonar Pro API | Generates hooks, analogies, citations, and quiz prompts | 
| Visual Layer | LangChain + Gemini 2.0 Flash | Creates visuals to reinforce learning | 
| Hosting | AWS | Full-stack deployment | 
🤖 AI Integration Details
Perplexity Sonar API: Real-Time Relevance
- Trained on live web data
 - Handles informal, meme-like prompts
 - Outputs structured developer-friendly JSON
 
Integration Flow:
- 🧱 Prompt templating via LangChain
 - 🧾 Output schema for hook, analogy, quiz, riddle
 - 🔍 Fallback JSON parser (
extract_valid_json) - 🧵 Used to generate Feed Cards and HookQuests
 
See more in Feed and Trending.
📂 Documentation Sections
Navigate through each key component of the system:
- 🖼 Frontend UI – Built with React Native for a buttery-smooth mobile experience.
 - 🧠 Backend Logic – FastAPI-based service for XP, hooks, and user flow.
 - 📰 Feed Generation – How we generate bite-sized learning cards from Sonar responses.
 - 📊 Trending Engine – How popularity scores drive the global trending page.
 
🧩 Engineering Challenges
Not Everything Was Plug-and-Play
- Hooks had to be carefully trimmed and cleaned to stay below 60 words.
 - Quizzes needed post-processing and logic to balance difficulty.
 - UI needed dopamine without distraction—especially for neurodivergent users.
 - Chain-of-thought logic was simulated via custom thread-like design.
 
🏆 Accomplishments
- ✅ AI-powered, real-time educational feed
 - 🎯 Designed for ADHD, anxiety, and cognitive load
 - 🧭 Story-driven “HookQuests”
 - 📱 TikTok-style UI for knowledge
 - 🧠 Turned raw LLM output into structured learning moments
 
📚 Lessons Learned
LLMs are powerful—but unpredictable.
Prompt engineering is a full-time job. We learned to debug and mold outputs for precision and emotional tone.
- Designing for neurodivergent users means optimizing for safety and clarity.
 - Gamification must be earned, not forced.
 - React Native + FastAPI = fast iteration.
 - AI text → structured logic = creative problem solving.
 
🔮 What's Next
- 🧠 Hook History & Recommendations
 - 👥 Social Learning & Study Squads
 - 🧪 Contextual Quiz Engine
 - 🎭 Mood-Based Hooking
 - 🌍 Offline Mode + Multilingual Support
 
🎥 Demo
🔒 Repository
Shared With:
- james.liounis@perplexity.ai
- sathvik@perplexity.ai
- devrel@perplexity.ai
- testing@devpost.com
🏷 Submission Info
Category: Education, Health, Fun and Creative
Bonus: Deep Research
Team: Aishwarya Jakka, Rahul Rao, Siddarth Karmokar
Main Tool: Perplexity Sonar API
📬 Contact
Created by:
- Siddarth Karmokar - siddkarmokar@gmail.com (backend)
 - Aishwarya Jakka – aishwaryajakka@gmail.com (ui, ux and management)
 - Rahul Rao - rahulyup@gmail.com (frontend)
 
📜 License
MIT License — Open for educational, non-commercial, and community use.