Portrait of Zhehao Zhang

Zhehao Zhang

About

I am a graduating student (also researcher) at East China University of Science and Technology. My research interests include computer organization and architecture, processor's architecture, Microcontrollers, articifial intelligence and application and software develop. I am passionate about retro computer, outdoor adventure and vibe codingin my spare time.

Education

Master of Science in Computer Architecture

Northwestern Polytechnical University, Xi'an, Shaanxi, China

2026.9 – 2029.6

* Recommended for graduate admission without entrance examination

Bachelor of Science in Computer Science and Technology

East China University of Science and Technology, Shanghai, China

2022.9 – 2026.6

Key Courses: Data Structures and Algorithms (93), Computer Organization and Architecture (94), Computer Networks (94), Database Principles (94), Operating System (CS) (93), Computer Architecture (94), Computer Graphics (92), Embedded Systems (96), Microcomputer Principles and Interface Technology (94), Natural Language Processing and Understanding (95), Probability and Mathematical Statistics] (96), Artificial Intelligence (92),

Required Courses: 3.77/4.00, Overall: 3.68/4.00, Average Score: 88.53

Rank: 2/107(2024-2025)

Research

Multi-Agent based RTL Code Generation and Verification System

2025.10 – Present

This project focuses on constructing a multi-agent standardized workflow for RTL code generation and verification. It introduces established RTL design methodologies and constrains the model generation process through structured prompt sequences covering functionality, timing, signals, and registers. By employing a pipeline-based multi-agent collaboration mechanism, complex design tasks are decomposed into verifiable subtasks. Furthermore, integrating the UVM methodology with industrial-grade verification tools, a verification-driven automated closed-loop system is established, enabling LLMs to perform code refinement and iterative optimization based on verification feedback. This approach enhances the reliability and engineering practicality of automated RTL generation, providing a new technical pathway for intelligent chip design.

  • Multi-Agent
  • Circuit Verification
  • AI Aided EDA
  • Circuit Design

One Student One Chip (YSYX)

2026.3 – Present

The teaching team of this project proposes the "Computer System Capability 3.0" processor chip talent cultivation goal. This approach takes open-source processor chips as the entry point, employs processor chip agile development methods as experimental means, and integrates courses from Computer Science (CS) and Electronic and Information Engineering (EE) through a coherent design. It highlights the integration of science and education as well as industry-academia collaboration, emphasizing both theory and practice. Through a teaching tape-out program, it achieves processor chip instruction on silicon, cultivating full-stack top-tier talents in the computer systems field. The core concept of "One Student, One Chip" — simply put — is "to enable a student to graduate with a processor chip of their own design." The goal is to lower the design threshold of processor chips through a teaching mechanism that balances theory and practice, allowing more students to participate in every stage of the processor chip design process in a full-flow manner.

  • IC Design
  • Digital Circuit Design
  • Computer Architecture
  • Pipeline Processor Design
  • Superscalar Processor Design

Automatic Depression Speech Recognition

2024.4 – 2025.5

This project focuses on researching and studying automatic depression recognition methods using speech signals over the past five years, based on speech signal processing and deep learning technologies. The scope includes, but is not limited to, datasets, methods, and models. Emphasis is placed on analyzing acoustic features of speech and mining depression-related speech behavior patterns. Neural networks are employed to extract temporal speech features and build corresponding models, as well as to convert speech data into quantifiable assessment metrics. The final outcome will be the formation and writing of a review paper.

  • Review
  • Dataset
  • Model

Carbon Data Origins – A Data-Driven One-Stop Emission Reduction System for Wastewater Treatment Plants

2023.4 – 2025.4

This project uses algorithmic models to accurately predict non-CO₂ greenhouse gas emissions from wastewater treatment plants, including but not limited to nitrous oxide. It verifies and corrects emission factor data for various treatment processes, converts monitoring data into regularized models, and builds an LSTM model using both online and offline methods to precisely predict non-CO₂ greenhouse gas emissions. Through dynamic feedback-based adjustment of emission parameters, a comprehensive greenhouse gas emission prediction scheme for wastewater treatment plants has been developed, which is currently in practical operation.

  • LSTM
  • Machine Learn
  • Time Series Analysis

Projects

APPLE ][ Computer replica

2026

This project restores the 1977 Apple II computer from scratch to the greatest extent possible, using modern microcontrollers (without FPGAs) to replace certain no longer available chips, such as the 2513. The final result is a system capable of running software originally supported by the Apple II.

Z80 Single-Board Computer

2025

This project designs a complete Z80 retro computer based on the Zilog Z80 processor. It uses 74-series chips throughout, with only the address decoding implemented using GALs (Generic Array Logic). The system supports full memory mapping and is capable of running the CP/M 2.2 operating system as well as related peripheral software.

Technologies: EasyEDA, Z80, Assembly Language, Computer Organization, PLD

RISC-V (RV32I) Processor

2026

This project designs a simple RISC-V single-cycle processor using FPGA and Logisim, including a complete implementation of the RV32I instruction set and basic peripheral management. All instructions have passed testing, and the processor is capable of running full C language programs and producing output results.

Technologies: Verilog, RISC-V, Processor

Linux Better Shell

2025

This project develops a BASH-like shell program using Rust and Go, capable of fully replicating BASH operations, including terminal process management, stream redirection, and event handling. In addition, based on the original functionality, a series of high-performance, memory-safe TUI visual operation components have been developed in Rust to facilitate ease of use.

Technologies: GoLang, Rust, Shell, Operation System

Intelligent Flipped Learning Platform Based on Large Language Models

2024

This project aims to leverage advanced technological methods to achieve intelligent learning, making the learning process more personalized, interactive, and efficient. By integrating large language models and innovative teaching models, the platform breaks away from traditional learning frameworks, allowing users to act as both students and teachers, thereby facilitating mutual learning and knowledge sharing. In addition to technical implementation, the platform also offers a range of distinctive features, including personalized learning plan generation, a points-based reward mechanism, and a learning marketplace ranking, motivating users to actively participate in the learning process. The introduction of these features not only enhances the fun and interactivity of learning but also provides users with greater motivation and rewards, thereby promoting the dissemination and sharing of knowledge.

Technologies: Java, Hibernates, TiDB, LLM

Database Management System and Kernel

2024

This project implements a complete relational database management system and its core engine in Java. It supports create, drop, modify, and query operations on databases and tables. Indexing is implemented using B+ trees and hash structures. The system also supports advanced features such as join queries, compound statements, and GRANT/REVOKE permission operations.

Technologies: B+ Tree, Java, Index, Storage, Database Kernel

Distributed Storage System (DSS)

2023.9 – 2023.12

A distributed storage system implemented in Golang, designed to provide reliable, scalable, and high-performance file storage services across multiple storage nodes. The system adopts a master-slave architecture where the metadata node manages the global namespace and file chunk locations, while multiple data nodes handle actual data storage and retrieval. The core design implements file chunking with configurable block sizes. The system features a unified RESTful API interface, supports concurrent upload, download, look through details with byte-range requests. The project emphasizes high availability, data durability, and linear scalability, aiming to provide an educational yet production-inspired distributed storage solution that enables hands-on experience with core distributed systems concepts including consistency, fault tolerance, and load balancing.

  • Golang
  • Distributed File System
  • File Chunking & Metadata Management
  • Replication & Erasure Coding
  • RESTful API

Carbon Emission Prediction and Strategy Planning Platform

2024

This project is a visual control platform developed based on the research project "Carbon Data Origins – A Data-Driven One-Stop Emission Reduction System for Wastewater Treatment Plants". It serves as an operational platform distributed to terminals.

Technologies: Diagram Design, LSTM, Deep Learning, PyTorch

Academic Affairs Intelligent Query System

2023

This project implements supplementary commonly used functions for the existing academic affairs resource query platform on campus. It integrates advanced features such as empty classroom query, classroom and teacher schedule inquiry, making it convenient for students to use when self-studying.

Technologies: GoLang, Gin, PostgreSQL

Intelligent Health Assessment and Management System Based on DenseNet

2023

In the post-pandemic era, the general public has become increasingly concerned about their health. However, due to the relative scarcity of medical resources, individual health needs cannot always be fully met. To address this issue, we have developed this intelligent health assessment and management platform using Django. The system designs multiple access roles, including general patients, doctors, and global administrators, with corresponding functions implemented for each role. It cleverly integrates health education, diagnosis, medicine procurement, order management, and other functions into one platform to enable intelligent health management, while supporting both online and offline medical activities. By leveraging algorithms such as the binary method and DenseNet, the system can diagnose and assess certain diseases. The platform is easy to operate, helps meet the health needs of a broader range of community members, alleviates the service pressure on doctors, and improves the overall user experience and service quality.

Technologies: DenseNet, Bisection, Full Stack, Computer Vision

Publications

A Review of Advances and Future Trends in Satellite Communication Systems [pdf]

Hao Xu*, Zhehao Zhang, Chengxiang Ge

Journal of Engineering Science & Applications, 2026, 1(2): 1-12

A Systematic Review of Speech-Based Depression Detection [pdf]

Zhehao Zhang, Haodong Shen, Xinyue Du, Sihan Jin, Yanlin Ju, DongDong Li*

软件著作权：碳排放预测与策略规划平台

登记号：2024SR2051943

Intern Experience

Backend Developer

NIO Corporation, Shanghai

2025.10-2026.1

Microsoft Learn Student Ambassador

Microsoft Corporation, Global

2024.10-2026.6

Deputy Manager

East China University of Science and Technology's IT Department, Shanghai

2023.4-2024.6

Awards & Honors

Skills

Programming: Python, Verilog, Java, C#, SQL, Golang, C

Tools & Frameworks: Django, Spring, ASP.NET, LangChain, Altium Designer, Quartus, Verilator

Technical Stack:Scientific Research, Network, Server, Develop, Management

Languages: Chinese (native, can read, can write, can speak, can understand), English (fluent, can read, can write, can speak, can understand)

Transcript

Undergraduate (East China University of Science and Technology, Bachelor of Science in Computer Science and Technology) English Version Chinese Version