My name is Johnny Jiang. I'm a recent new Grad of University of Toronto in Canada.
- 🧐 Interested in full stack development. Recently focus on MERN Stack.
- ✏️ React.js / HTML / CSS / Django REST / Python / SQL
- 🎓 Honours Bachelor of Science in Computer Science GPA 3.94/4.00
- 📚 Currently learning Node.js / MongoDB / Kubernetes
- 💭 Ask me anything at Discussions!
Check out my projects!
✨ Astra - Python / SQLAlchemy
- A GUI-based application for QEYnet Inc., enabling their employees and customers to read and interact with satellite telemetry data and manage tag-sensitive alarms.
- Designed and engineered a robust database using SQLAlchemy for efficient data retrieval and storage across multiple devices, alarms, and telemetry data.
- Implemented reliable and scalable file I/O mechanisms for reading custom telemetry files, ensuring data persistence across sessions and seamless integration with diverse device configurations.
- Conducted comprehensive testing within a CI workflow, ensuring data integrity and application reliability.
- Collaborated within a full Agile development cycle using Scrum and Kanban methodologies via Notion, and perfectly accomplished all the partner’s requirements.
🌱 Petpal - HTML / CSS / Django REST / JavaScript / React.js
- A comprehensive pet adoption platform that enables pet seekers to search and apply for pets, while allowing pet adopters to establish shelters, list pets, and manage adoption approvals.
- Frontend - Developed a dynamic and responsive user interface using React, incorporating React Hooks for efficient state management, React Router for seamless client-side routing, and intuitive navigation features.
- Backend - Constructed a robust RESTful API with the Django REST Framework, utilizing models and serializers alongside CRUD Views to handle all data operations effectively. Implemented token-based authentication using JWT to secure user data.
🔭 Empirical Study on BERT-based model - PyTorch / Deep Learning
- Developed a BERT-based salary prediction model for Data Analyst on real job datasets (information with texts after preprocessing) with an 84% test accuracy.
- Conducted an empirical study that compares the performance among the augmented pre-trained BERT model, RNN model and the gradient-boosted decision tree (GBDT/baseline) on tabular data.
