Hello, visitor! 👋

😀Hi! I’m Flynn, a first year graduate student from Santa Clara University(MS, Computer Science and Engineering). I received my bachelor degree at the South China University of Technology(BS, Information and Computing Science)

I am a C++/Python/Java developer, experienced in Backend development & Server/Client development.
My interests include backend development, LLMs, computer graphics, and OS&Network.

Trying to do better🚀

🎓Education

Santa Clara University

MS, Computer Science and Engineering, College of Engineering
Sep. 2024 – Expect Jun. 2026

GPA: 3.85
Coursework: Computer Architectures, Design and Analysis of Algorithms, Operating Systems, Computer Graphics, Data Structures, Distributed Systems, Logic Design

South China University of Technology

BS, Information and Computing Science, College of Mathematics
Sep. 2020 – Jun. 2024

GPA: 3.43, Rank: 19/63
Coursework: Mathematical Analysis, Advanced Algebra, Numerical Solutions of Differential Equations, Numerical Analysis, Probability, Operations Research, Machine Learning, Computational Intelligence

👔Professional experience

ZWSOFT

C++ Development Engineer
Guangzhou, China, Jun. 2023 – Oct. 2023

Designed a high-performance, low-latency memory pool for 3D engineering software using STL and Boost.
Implemented the Singleton pattern and a simple segregated storage strategy, providing custom malloc/free and new/delete operations for vertex, line, and polygon objects, accelerating geometry generation and export by 40%.
Utilized perfect forwarding with templates to eliminate unnecessary copies. Ensured thread safety with std::atomic for lock-free pointer operations and std::mutex for secure block allocation and recycling.
Optimized memory usage by reducing internal fragmentation via memory alignment and external fragmentation through contiguous block preallocation. Used placement new for object reuse on pre-allocated blocks, reducing memory footprint by 15% across 3D geometries.

📚Research experience

ECG Frequency Classification Based on Deep Neural Networks

Research Assistant - Jan. 2023 – Nov. 2023
Prof. Charlie (Li) Xia, LabX, College of Mathematics, SCUT

Read ECG classification-related papers, conducting experiments for validation.
Trained and Integrated ECG frequency classification models by using Transformers and CNN.
Published CTRhythm: Accurate Atrial Fibrillation Detection from Single-Lead ECG by Convolutional Neural Network and Transformer Integration.

Natural Language Processing Technology Research

Research Assistant - Jan. 2023 – Jun. 2023
Prof. Qianli Ma, Machine Learning&Data Mining Lab, College of Computer, SCUT

Read papers and techniques related to NLP and Named Entity Recognition (NER problem).
Reproduced and conducted experiments on BOND with dataset CoNLL and OntoNotes, providing benchmark data.

JPEG Image Encryption Algorithm Research

Research Assistant - Apr. 2022 – Apr. 2023
Prof. Junhui He, College of Computer, SCUT

Researched image encryption algorithms, reproduced image encryption algorithm based on paper.
As a full-stack developer, developed a mobile image encryption app based on the Android platform and reproduced image encryption algorithm.

🛠️Project experience

Distributed KV database based on Raft consensus algorithm

Dec. 2024 – Present
Framework: C++, Boost, STL, Muduo, protobuf

Implemented log replication and leader election for the Raft consensus algorithm, leveraging thread pool to manage heartbeat and election tasks, ensuring log consistency and cluster stability.
Designed and developed an RPC framework with Protobuf, enabling remote procedure calls and serialized data transmission between Raft nodes.
Built a skiplist-based key-value database for high-performance data storage and retrieval.

MathMind: LLM-powered Math Helper – Scan & Solve Instantly

Dec. 2023 – May. 2024
Framework: Python, Java, Kotlin, SQL, Flask, PyTorch, Android Studio, Docker

Designed and implemented scalable LLM-powered math-solving microservices in Python, PyTorch, and Flask, with a mobile app in Java and Kotlin.
Developed RESTful APIs and WebSocket connections. Decoded and preprocessed Base64-encoded images using OpenCV and Pickle, and integrated Transformer-based LaTeX OCR models for text and formula recognition in scan services.
Fine-tuned ToRA-7B and Llama3 on algebra and calculus datasets using supervised fine-tuning (SFT), deployed models using TensorRT-LLM, and integrated remote GPT and Gemini APIs for solving services.
Built Docker images, managed multi-version Python environments using Miniconda for subprocess calls. Orchestrated deployments with Docker Compose and automated deployment tasks using Shell/Bash scripts.
Developed an Android app using MVVM architecture, leveraging Kotlin Coroutines for asynchronous tasks, Handlers for UI updates, and SQLite for local persistence of historical data.

A High-Concurrency C++ Server Library Based on Muduo Library

Apr. 2023– Jul. 2023
Framework: C++, Muduo, Boost, STL

Implemented a high-concurrency server using non-blocking I/O, multiplexing, and the Reactor pattern, inspired by Muduo and Boost.
Developed core components such as EventLoop, Poller, and Channel to handle event-driven loop listening, request dispatching, and asynchronous event processing.

📖Publication

CTRhythm: Accurate Atrial Fibrillation Detection from Single-Lead ECG by Convolutional Neural Network and Transformer Integration

Authors: Zhanyu Liang, Chen Yang, Zhengyang Yu, Yinmingren Fu, Bozhen Ren, Maohuan Lin, Qingjiao Li, Xuemei Liu, Yangxin Chen, Li C. Xia

👨🏻‍💻Technical Skills

Programming Language: C/C++, C#, Java, Python, Golang, JavaScript, HTML, SQL, Shell/Bash, CUDA
Deep Learnging: Pytorch, TensorFlow, Scikit-learn, PaddlePaddle, Matplotlib, Pandas
Backend Development: C++(Boost, STL, Muduo, Linux), Java(Spring Boot, Nginx, Netty, Kafka)
Tools: Git, Docker, Kubernetes, Android Studio, AWS, gdb, cmake, TortoiseGit, Mercurial, Github Actions
Databases: MySQL, SQLServer, Redis, PostgreSQL, MongoDB
Concepts: Reactor, RPC, Raft

In my daily life

👨‍💻Programming, Animation, 🎮Viedo Games, 🤖Gundam Model/Animation/Movie, JPOP, 🎥Moive and 🚶‍♂️strolling around