Skip to content
View GioiaZheng's full-sized avatar

Block or report GioiaZheng

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
GioiaZheng/README.md

Gioia Zheng

B.Sc. in Applied Computer Science & Artificial Intelligence Sapienza University of Rome · Italy

Email · LinkedIn · LeetCode · TryHackMe · GitHub


Overview

I am an undergraduate student in Applied Computer Science and Artificial Intelligence at Sapienza University of Rome, focused on building scalable machine learning systems and retrieval-driven applications.

My work sits at the intersection of:

  • Information Retrieval (Search & Ranking)
  • Machine Learning Systems
  • Backend Engineering

I am particularly interested in designing end-to-end pipelines that connect retrieval, ranking, and model inference into production-oriented systems.


Selected Projects

MS MARCO · BM25 + FAISS + Transformer Reranking

  • Built a hybrid retrieval pipeline combining sparse (BM25) and dense (FAISS) search
  • Implemented transformer-based reranking to improve top-k relevance
  • Designed modular system for indexing, retrieval, and ranking
  • Explored trade-offs between recall, latency, and scalability
  • Structured for extensibility (plug-in retrievers / rerankers)

Go · REST APIs · Concurrency

  • Developed backend system in Go with concurrent request handling
  • Implemented RESTful APIs for messaging and session management
  • Designed lightweight architecture for real-time communication
  • Focus on performance and system reliability

CNN + LSTM · Sequence Modeling

  • Built OCR pipeline using CNN + LSTM architecture

React Native · Backend Integration · UX Design

  • Developed a cross-platform mobile app for food discovery in Italy
  • Designed user interface tailored for international users navigating local menus
  • Integrated backend APIs for data-driven recommendations
  • Iteratively improved UX based on user feedback and testing

Current Work

  • Developing Retrieval-Augmented Generation (RAG) pipelines
  • Improving retrieval quality and ranking performance
  • Studying system design for ML-backed applications
  • Practicing algorithms and systems (LeetCode)
  • Hands-on security labs (TryHackMe)

Technical Skills

Programming

  • Python · Go · JavaScript · SQL

Machine Learning & Retrieval

  • PyTorch · Scikit-learn · OpenCV
  • BM25 · FAISS · RAG pipelines

Backend & Systems

  • REST APIs · Docker · SQLite · Git
  • Scalable system design (fundamentals)

Frontend

  • Vue.js · React Native

Security

  • TryHackMe (Pre-Security · Intro · Labs)

Interests

  • Retrieval systems & ranking pipelines
  • Scalable ML applications
  • Backend architecture
  • Reliable and secure systems

Languages

  • English (Fluent)
  • Italian (Native)
  • Chinese (Native)
  • German (Beginner)

Building systems that combine learning-based methods with solid engineering.
🧠 Mensa Member (Top 2% IQ Society)

Pinned Loading

  1. msmarco-genqa msmarco-genqa Public

    RAG-based search question answering system on MS MARCO with BM25 retrieval, FAISS indexing, and transformer reranking.

    Jupyter Notebook 16

  2. go-chat-system go-chat-system Public

    Full-stack chat service built with Go, SQLite, and Vue.js featuring authentication, persistent messaging, and REST APIs.

    Vue 4 1

  3. handwritten-ocr-system handwritten-ocr-system Public

    Deep learning-based handwritten OCR system using CNN-RNN-CTC with evaluation via CER/WER metrics.

    Python 8

  4. CiboCompass CiboCompass Public

    Full-stack mobile app (React Native + Go) with REST API, offline-first sync, and cultural-aware food exploration

    JavaScript 7

  5. Leetcode-Solutions Leetcode-Solutions Public

    Structured Python implementations of LeetCode problems with automated tracking, algorithm notes, and reusable patterns for interview preparation.

    Python 16 3

  6. cybersecurity-learning cybersecurity-learning Public

    Python 1