About Myself
Hi, I am Mingrui Xiao
Nice to meet you! I am a Software Engineer focused on React + TypeScript, Node/Java, and cloud-native delivery. I like shipping products with clean UX, solid APIs, and reliable DevOps.
Skills
Full-Stack Development & Cloud Architecture
I build scalable, secure, and user-centric web applications from frontend to cloud infrastructure.
Backend Systems, APIs & Data Engineering
I design reliable backend systems that power data-driven and distributed applications.
Generative AI & Machine Learning
I build practical GenAI solutions that enhance productivity and decision-making.
Projects
Farm Management CMS
- Built a centralized Farm Management CMS managing 2000+ assets, activities, and bookings using React (TypeScript) and Azure Functions.
- Designed core modules including Farm, Asset, Activity, Booking, Feature Tag, and User Management with Prisma and Azure SQL.
- Implemented secure role-based access control using Microsoft Entra ID.
- Developed accessible, responsive UI with MUI and global theming.
- Integrated Azure Blob Storage and optimized backend API performance.
- Built CI/CD pipelines reducing deployment time from 10 min → 2 min.
Food Delivery Platform
- Led a 7-member cross-functional team to build a food delivery platform serving 3,000+ users in the Nagoya region.
- Designed, developed, and launched a full-stack food delivery application using C#, .NET, and Microsoft SQL Server.
- Improved order-processing efficiency by 50% through backend optimizations and database indexing.
- Reduced rider assignment time by 40% by optimizing dispatch and routing logic.
- Built an Admin Back-Office Management System to manage riders, restaurants, orders, and attendance, significantly improving operational efficiency.
- The platform was later acquired by HungryPanda, a global food delivery company.
Smart Trainer (Pose + LLM)
- Trained a PyTorch pose estimation model achieving 92% detection accuracy for major joints in standard fitness movements.
- Implemented real-time inference pipeline reducing frame processing latency to < 80ms.
- Using flask to connect python backend to react frontend.
- A smart agent function to analysis and generate plans to trainee, using LangChain to link ChatGPT API.
- Using springboot framework connect PostgreSQL database to store record and exercise dataforagentto analysis.
Research: GAN-Based Image Generation
- Conducted research on Generative Adversarial Networks (GANs), focusing on feature-controlled image generation.
- Improved StyleGAN2 architecture to enhance image quality and feature consistency.
- Designed and trained deep learning models using Python in a Linux environment.
- Leveraged CUDA acceleration with PyTorch and TensorFlow to optimize model training performance.
- Published a research thesis in Japan, demonstrating practical advancements in GAN-based image synthesis.
Experience
Full Stack Developer
- Optimised and enhanced the school’s existing website, delivering improvements across front-end UI/UX and back-end data processing to improve system stability and user experience.
- Designed and developed a new CRM system, integrating legacy school systems with Google Admin to centrally manage data for 400+ students and administrative staff, significantly improving data consistency and operational efficiency.
- Implemented modern LLM-based features within the new system to automate workflows, support intelligent data processing, and enhance administrative productivity.
- Collaborated closely with administrative teams to analyse requirements, streamline student–teacher data management processes, and reduce manual workload.
Nov 2025 — Present
Master of Information Technology and Master of Information Technology Management
- Master of Information Technology and Master of Information Technology Management.
Jul 2023 — Jul 2025
Software Engineer
- Led a 7-member cross-functional team to build a food delivery platform used by 3,000+ local customers in the Nagoya region.
- Designed, developed, and launched a food delivery application, QING QI SU DA, using C#, .NET, and Microsoft SQL Server, which was later acquired by HungryPanda, a major global food delivery platform.
- Increased order-processing efficiency by 50% through backend optimizations and database indexing using Microsoft SQL Server.
- Reduced rider assignment time by 40% through optimized dispatch logic.
- Built an Administrator Back-Office Management System to manage rider profiles, restaurant partners, delivery orders, and time/attendance tracking, enhancing operational efficiency.
Mar 2020 — Dec 2020
Volunteer | Tutor
- Teaching high and middle-school students Japanese and Mathematics.
Sep 2017 — Jun 2023
Bachelor of Mathematics
- Bachelor of Mathematics.
Apr 2019 — Mar 2023
Contact
Email: shawnxiao.jobs@email.com