Projects
Personalized Twitch Resources Recommendation Engine
📅 Dec 2023 – Mar 2024
Tech Stack: Java, Spring Boot, React, MySQL, Redis, REST API, Docker, AWS, CI/CD
- Designed and implemented a Spring Boot backend with MySQL, optimizing query performance and reducing response time by 35% (200ms to 130ms) through indexing and caching
- Optimized a collaborative filtering recommendation algorithm, reducing processing time from 11.8s to 9.7s for 10,000 videos and 100 users, achieving a 17.8% efficiency boost
- Deployed the system using Docker and AWS, automating CI/CD pipelines via GitHub Actions, cutting deployment time from 30 min to 13 min, ensuring seamless integration and continuous updates
E-commerce System
📅 Jul 2023 – Aug 2023
Tech Stack: Java, Spring Boot, Vue.js, MySQL, Redis, Elasticsearch, REST API, Docker, AWS
- Led backend development of a full-stack e-commerce system, designing and implementing the shopping cart, order management, and authentication modules using Spring Boot & MySQL
- Optimized system performance by integrating Elasticsearch for search (-42% latency) and Redis for caching (-44% backend response time) under high traffic
College Admission Management System
📅 Aug 2022 – Sep 2022
Tech Stack: Java, Spring Boot, MySQL, JDBC, Redis, REST API
- Developed backend for a college admission management system, implementing student registration, application submission, and result management with Spring Boot & MySQL
Lossy Audio Compression Algorithm Design
📅 Oct 2024 – Dec 2024
Tech Stack: Python, NumPy, SciPy, Huffman Coding, DCT, Signal Processing
- Led a two-person team to develop a lossy audio compression algorithm, reducing WAV file size while ensuring high audio fidelity under a specified distortion limit
- Collaborated on optimizing Discrete Cosine Transform (DCT) and Huffman coding, achieving an 8.81:1 compression ratio and a final compressed size of 74,057 bytes, while maintaining an MSE of 0.0048
- Coordinated performance testing and refinements, ensuring the algorithm met all constraints, with a decompression time of 0.17s for real-time playback feasibility