Hongshen Zhang

Email: zhanghongshen716@gmail.com

Education

Nanyang Technological UniversityJan, 2025 - Jun, 2026

Master of Science in Machine LearningSingapore

Southwest Jiaotong UniversitySep, 2018 - Jun, 2022

Bachelor of Engineering in Computer ScienceChina

Work Experience

ByteDanceJul, 2022 - Apr, 2023

Backend EngineerChina

  1. Old Lark System Retirement & Data Migration
    Developed and tested data migration scripts for the retirement of the old Lark system. Successfully migrated data from 300 enterprises , covering 20 data tables with a total of 20 million rows. The process was completed without incidents and reduced system maintenance costs by 30%.
  2. Technical Owner of Lark Headhunting Project
    Led technical planning and solution design. Participated in requirement analysis, resolved 10+ online issues, optimized troubleshooting workflows, and maintained zero incidents throughout the development cycle.
  3. Unit Testing Optimization
    Led improvements to unit testing across team projects. Enhanced coverage for legacy code and ensured all incremental code changes maintained over 70% test coverage, raising overall project coverage by more than 10%.

Internship Experience

AmazonJun, 2025 - Aug, 2025

Software Development Engineer InternChina

  1. Designed and implemented an efficient RAG workflow using Python, Java, AWS, enabling large-scale document knowledge retrieval and intelligent responses.
  2. Built a conversational interface with React Native, optimized specifically for input/output-constrained environments like Kindle, ensuring smooth and low-latency user interactions.
  3. Developed core native modules in C++ to handle network requests and memory management, ensuring high efficiency and stability even on resource-constrained devices.

Project

ModelCraftJan, 2024 - Now

Tech Stack: Swift, SwiftUI, SwiftData, SwiftSoup, Ollama

  • Implemented a ReAct-based Agent engine to enable multi-step task orchestration, allowing the LLM to autonomously invoke local tools and significantly enhancing complex problem-solving capabilities.
  • Developed a Rolling Summary mechanism to dynamically compress long conversation histories, effectively resolving Token overflow while maintaining context consistency and long-term memory.
  • Optimized RAG workflows with a local vector database and multi-modal parsing (PDF, Audio, Images, etc.), reducing model hallucinations and improving response precision through semantic retrieval.

    • Sentiment Analysis for Financial NewsMar, 2025 - Mar, 2025

    Tech Stack: Python, PyTorch, Huggingface Transformers

    1. Preprocessed financial news data, performed label encoding, tokenization, and dataset splitting.
    2. Supervised fine-tuned a pre-trained BERT model using HuggingFace Transformers, enhanced with LoRA for efficient parameter tuning.
    3. Applied CrossEntropy loss with L2 regularization and early stopping to mitigate overfitting and improve generalization.
    4. Achieved 86% classification accuracy, outperforming traditional ML baselines.

    Rednote-like social appDec, 2024 - Jan, 2025

    Tech Stack: Spring Boot, Spring Cloud, Nacos, MySQL, Redis, Kafka, Elasticsearch, Minio, Freemarker

    1. Implemented scheduled publishing for articles and full-text search functionality.
    2. Used Freemarker template engine to render article detail pages, improving page load performance.
    3. Performed offline article scoring and leveraged Kafka Streams for real-time score updates.
    4. Implemented multi-level rate limiting, including Nginx rate limiting, Gateway rate limiting, and microservice-level rate limiting, to enhance service stability.

    Skills

    1. Proficient in Java, Python, Typescript, React.js.
    2. Experienced with MySQL, Redis, Kafka and RocketMQ.
    3. Skilled in developing backend services using Spring Boot and Spring Cloud.
    4. Experienced with Pytorch, Scikit-learn, Transformers.
    5. Familiar with CNN, RNN, LSTM, Transformer architectures.
    6. Experienced with RAG and fine-tuning LLMs, including LoRA, Prompt-tuning, etc.
    7. Familiar with Pytorch, Scikit-learn, Transformers, Numpy, Pandas, etc.
    8. Familiar with AWS.

    Honors

    First prize, Sichuan Province, China Undergraduate Student Market Survey and Analysis Competition Apr, 2021

    Second prize, WeChat Mini Program Application Development CompetitionDec, 2020

    First prize, Sichuan Province, China Undergraduate Mathematical Contest in Modeling Sep, 2020

    Third prize, National University Students Electrical Math Modeling Competition Jul, 2019

    Third prize, May Day Mathematical Contest in ModelingJun, 2019

    Last Updated: 19 Jan, 2026