I received my Ph.D. degree in Electronic Science and Technology from Beijing Institute of Technology in 2025 🎓. My academic research centered on 3D medical image analysis, focusing on unsupervised/semi‑supervised brain MRI lesion segmentation with deep learning.
After graduation, I worked on large‑model‑driven search & recommendation systems. My industrial experience covers multi‑modal large language models (VLMs) upgrade for large-scale search engines, e‑commerce live‑video recommendation, and LLM‑based generative retrieval for advertising systems.
I am also passionate about low‑level system & graphics programming 🛠️. I build personal open‑source projects including CPU/GPU renderers and lightweight deep‑learning frameworks implemented in C++ / CUDA.
- 📫 Reach me: chenghao1652@126.com
Medical Computer Vision, Multi‑modal LLM, Recommendation & Search Systems, High‑performance Computing, Graphics Rendering
- 💻 Programming some interesting stuff.
- 🎯 Rigid/soft body simulation using C/C++: XPBD.
- 🌄 Computer graphics
- 🌄 Traditional: ray/path tracing, GI, PBR (physically based rendering).
- 🌄 Modern: NeRF, DDPM.
- 🌄 Parallel computing with CUDA C++ & NVRTC.
- 🎥 3D modeling: 3Ds MAX, Blender.
- 💖 I’m also fascinated by pixelated art styles and old-school graphics.
- 🔨 SGL (Software Graphics Library): Implementing my own OpenGL on the CPU, with
- 🔮 Fully customizable vertex and fragment shader support
- 🔮 Skeletal animation support
- 🔮 Decent rasterization performance (~2ms for 320x240 resolution with 1000 triangles)
