My name is Tianyi Huang (Tony), and I'm a student with a strong passion for AI and app development. My goal is to leverage these technologies to make a meaningful impact on the world and improve lives. Aspiring to be a leader in the world of AI, I'm continuously finding ways to keep innovating AI into everyday solutions.

I have experience with Python and frameworks like PyTorch and TensorFlow, web development using React and Node.js, and mobile application development with MIT App Inventor. My research focuses on topics such as large language models (LLMs), neuro-symbolic AI, early AI education, and various other AI techniques (RNNs, ensemble learning, etc.). When I'm chilling, I enjoy playing tennis, reading, and classical singing.

• Founder & President, App-In Club: A global student-led non-profit that offers FREE and OPEN access to cutting-edge information on AI and apps, empowering students of all ages to become AI and app innovators to make a positive impact on the world.
• Student Ambassador, App Inventor Foundation: As the first student ambassador, my goal is to help expand the foundation's mission—to empower people across the globe to create apps that improve their lives and uplift their communities. I hope to inspire others to create meaningful app solutions and contribute to a future where every student can utilize technology for positive change.
• Research Intern, MathGPT: I conduct research in ways to enhance LLMs, developing agents capable of advanced mathematical operations that generalized models like GPT-4o may struggle with.

• Ed Box: An app that serves as a platform to connect students for the purpose of redistributing educational materials.
• Snap Cook: An app designed to combat food waste by transforming excess ingredients into instant recipes through a simple photo.
• Scribe AI: A project that uses AI to replicate individual handwriting styles with a physical machine designed to write with a real pen.
• MDRNN for Polyphonic Music Generation: Utilizing a Multi-Dimensional Recurrent Neural Network (MDRNN) to generate new polyphonic music, aiming to overcome traditional limitations of generating beyond training data and avoiding repetitiveness.
• Ensemble ML Approach for Early Detection of HABs: Using ensemble learning, combined with data augmentation and multi-agent systems, to enhance the early detection of harmful algal blooms (HABs).

Feel free to contact me on GitHub, LinkedIn, or via email by clicking the buttons below!