Research & Engineering Projects

Selected projects demonstrating systems-building, algorithm implementation, and research prototyping. Full project list available on GitHub.

VocoType-linux

High-performance offline Chinese speech recognition system for Linux desktop environments

My Contributions:

  • Designed and implemented end-to-end pipeline: audio capture, VAD, ASR model inference, and system integration
  • Optimized for low-latency offline operation without cloud dependencies
  • Built user-friendly configuration system for model selection and input method integration

Stack: Python, PyTorch, ALSA, Linux input methods

GitHub·Details
VocoType-linux demo screenshot

Retrieval-Augmented Generation (RAG) System

Full-stack question-answering and document summarization system with LLM backend

My Contributions:

  • Fine-tuned LLaMA model for domain-specific Q&A tasks
  • Implemented vector database indexing pipeline for efficient document retrieval
  • Built Vue.js frontend and Spring Boot backend with RESTful APIs

Stack: LLaMA, FAISS, Vue.js, Spring Boot, Python

GitHub

Autograd Engine & Neural Network in C++

From-scratch implementation of automatic differentiation and neural network library

My Contributions:

  • Reimplemented PyTorch-style autograd with computational graph construction and reverse-mode differentiation
  • Built fully connected neural network layers, activation functions, and optimizers (SGD, Adam)
  • Validated against standard benchmarks to ensure numerical correctness

Stack: C++17, Modern template metaprogramming

GitHub

Gaussian SPH Fluid Simulation

3D fluid simulation combining Smoothed Particle Hydrodynamics with Gaussian splatting rendering

My Contributions:

  • Integrated DFSPH (Divergence-Free SPH) constraints for incompressible fluid behavior
  • Implemented anisotropic Gaussian rendering for realistic surface appearance
  • Optimized spatial hashing and neighbor search for real-time performance

Stack: C++, CUDA, OpenGL

Course project: Computer 3D Graphics and Deep Learning (NVIDIA collaboration)

Mobile Robot Pose Estimation

Point cloud-based localization module for robotic manipulation tasks

My Contributions:

  • Implemented multi-method point cloud denoising pipeline (statistical outlier removal, voxel downsampling)
  • Integrated ICP-based pose refinement for robust localization under sensor noise
  • Improved robustness for mobile manipulation in dynamic environments

Stack: Python, PCL, ROS

Course project: AI and Robotics for Mobile Robot Manipulation (Boston Dynamics collaboration)

Operating System Development

Teaching-oriented OS kernel with core subsystems (scheduler, memory manager, file system)

My Contributions:

  • Implemented process scheduling (round-robin, priority-based), virtual memory management (paging, demand paging)
  • Built simple file system with inode-based storage and basic POSIX system calls
  • Completed both NJUOS and MIT xv6 lab series for hands-on OS concepts

Stack: C, x86 Assembly, QEMU