Numo
Co-developed with a fellow student, Numo is a large-scale language-learning platform designed to assist users in becoming polyglots. Heavily inspired by Duolingo, it integrates local AI models for offline speech-to-text (STT), text-to-speech (TTS), and LLM text generation. Generative AI layers track learner behavioral patterns to dynamically optimize and personalize lesson curriculums. The application features a unique visual aesthetic with interactive Three.js 3D backgrounds. While active development has paused to focus on a dedicated Chinese language learning tool, it stands as a robust prototype covering core linguistic needs.
About This Project
Building an engaging, adaptive language assistant that uses local first AI pipelines to customize curricula based on behavioral learning patterns.
Designed a Tauri-wrapped desktop application combining local speech models (STT/TTS) and LLM engines with interactive Three.js graphics. Built tracking algorithms to customize student learning paths.
Created a highly optimized, visually striking assistant that maps professional curriculums. Active development was completed to apply these insights to a specialized Chinese learning sub-tool.
AI Systems & Desktop Developer
2026
Public
Personal
Technology Stack
Project Story
Building an engaging, adaptive language assistant that uses local first AI pipelines to customize curricula based on behavioral learning patterns.
Designed a Tauri-wrapped desktop application combining local speech models (STT/TTS) and LLM engines with interactive Three.js graphics. Built tracking algorithms to customize student learning paths.
Created a highly optimized, visually striking assistant that maps professional curriculums. Active development was completed to apply these insights to a specialized Chinese learning sub-tool.
Insights & Takeaways
Highlights
- Jointly developed language-learning application utilizing local STT, TTS, and LLM engines.
- Incorporates dynamic curriculum optimization based on generative AI behavior modeling.
- Features interactive 3D visual environments built using Three.js.
Challenges
- Orchestrating high performance local AI runtimes alongside 3D rendering (Three.js) on standard desktop environments.
- Mapping professional curriculum baselines into automated behavioral-adaptation scripts.
Lessons Learned
- Studying existing learning methodologies and using professional curricula as a baseline ensures the tool matches real educational standards, though modular development is key to final polish.
Related Work
Case Study
Academia Plus
C++ Qt prototype simulating students, teachers, courses and exams in a school management system.
Case Study
Auto Shutdown
Rust + React desktop app to automate PC shutdowns by timer, schedule or idle state.
Case Study
Azkari
Lightweight desktop app that displays Azkar reminders based on time of day, built with Tauri, SolidJS, TypeScript and Rust.
