📖About Me

🎓Education

• 2015.09 - 2019.06, as an Undergraduate student at Nanjing University of Science and Technology

📚Publications & Patents

💼Work and Project Experience

2024.09 - Now - Algorithm Engineer - Huawei 2012 Noah's Ark Lab, China

Responsibilities include:

• Participated in the development of denoising, smoothing, and fusion algorithms for scanning electron microscope (SEM) images, including traditional methods for contour extraction, contour alignment, contour averaging, and contour overfitting

• Fine-tuned the Segment Anything（SAM） model for contour localization and coarse extraction on SEM images
  

2021.11 - 2023.09 - Software Development Engineer - CMB Network Technology (Shenzhen) Co., Ltd., China

Responsibilities include:

• Managed backend and frontend modifications, along with routine vulnerability maintenance for corporate current account systems.

• Led the redesign and development of frontend and backend systems for corporate current account closures, covering CNY, cross-border CNY, and foreign currency accounts.

• Contributed to core development of corporate account opening systems (CNY, cross-border CNY), payment token management systems, corporate online banking systems, and linked payment systems.

• Handled database optimization, SQL performance tuning, backend refactoring, and architecture design for account opening and closure systems.

• Participated in multiple infrastructure reconstruction projects

2020.04 - 2021.04 - Software Development Engineer - Qianhai Hecheng Information Technology Co., Ltd., Shenzhen, China

Responsibilities include:

• Participated in the daily maintenance and feature development of the Intelligent Medical Emergency Platform and Mobile Emergency App

• Conducted on-site development at hospitals, integrating with HIS, LIS, PACS, RIS, EMR, and the 120 emergency dispatch system

• Assisted in the deployment, network configuration, and launch of emergency substation systems for 11 hospitals in Jiangmen, resolving technical issues during the process

🏆Honors & Awards

• 2018.12 - Excellent Student Scholarship - Nanjing University of Science and Technology Technology Consulting and Development Company, Jiangsu Granular Science Society

• 2018.10 - "Three Good Student" Title - Nanjing University of Science and Technology

• 2018.10 - First Prize - Jiangsu Province College Students' Granular Science Innovation Competition, Jiangsu Granular Science Society

• 2018.05 - Second Prize - 27th Innovation Cup College Student Extracurricular Academic and Technological Works Competition

• 2018.04 - First Prize - 2nd National University Science Popularization H5 Production Competition - China Science Daily

• 2017.12 - Ningbo David Medical Devices Co. Scholarship - Ningbo David Medical Devices Co.

• 2017.10 - "Three Good Student" Title - Nanjing University of Science and Technology

• 2017.09 - First Prize - Jiangsu Province College Students' Granular Science Youth Innovation and Entrepreneurship Competition - Jiangsu Granular Science Society

🎯Campus Leadership Experience

• 2017.06 - 2018.06 - Vice-chair, New Media Center, Youth League Committee, Nanjing University of Science and Technology

• 2016.06 - 2017.08 - Head of Publicity Department, School of Environmental and Biological Engineering, Nanjing University of Science and Technology

🤖LLMs Experience

Vision Detection and Segmentation Model：

• YOLO11 —— A neural network model for video and image segmentation and target detection

• segment-anything: A neural network focusing on image and video segmentation based on transformer

• FastSAM: Another neural network based on CNN backbone focusing on image segmentation

• MobileSAM: A nerual network based on segment anything (SAM) using light vit image encoder

• sam-hq: A nerual network based on segment anything (SAM) using external HQ-Output Token to keep high-quality segmentation

• Grounded-Segment-Anything: A workflow-based segmentation architecture using Grounded DINO for target detection and followed by SAM for segmentation

Image Generation Model：

• stablediffusion: High-Resolution Image Synthesis with Latent Diffusion Models

Voice Model：

• SenseVoice —— A Multilingual Voice Understanding Model

Common Model Tools & Services：

• Dify —— an open-source LLM app development platform

  • Developed various agent applications using Dify, integrated with models such as Qwen2.5 and ChatGPT, you can find it in my Dify website.

• LLama Factory —— Easily fine-tune 100+ large language models with zero-code CLI and Web UI

• Ollama —— A service for managing, quantifying, and running LLM models

  • Performed supervised LoRA fine-tuning on the Qwen2.5 model using a custom dataset, deployed it on Ollama, and integrated it with Dify for execution

• Xinference —— Another service for managing and running LLM models

🤝Contribution PR

Dify - Project：

• add the function of left folding in the exploration column #11368

🛠️Technology Stack

I am skilled in the development and deployment of project code engineering, with relevant experience in full-stack development. Below are the key technologies I commonly use:

• Java：spring、spring boot、spring cloud、spring gateway

• front-end-stack：vue2、vue3、nodejs）

• python：Numpy、Pytorch、Transformers

• database：oracle、mysql、postgresql、redis、Elasticsearch、MongoDB

• message middleware：rabbitMQ、rocketMQ、kafka
  

👥 Visitors