HELLO! MY NAME IS

SHELDEN SHI

I am a student at UC Berkeley stuyding Computer Science and Cognitive Science. I am passionate about Software Development, Cognitive Neuroscience, and Machine Learning.

Experiences

  • Roche Diagnostic 05/2021 - present

    Automation Engineer Intern

    ● Implemented standalone automation tools and data pipelines
    ● Used Python, Pandas and Tkinter library
    ● Increased consultants’ workflow efficiency by 500%
  • Thermaquil 02/2021 - 05/2021

    Consulting Project Manager | Software Developer

    ● Led a team of 5 undergrads to develop and design a embeded system UI for physicians and patients to control implantable nerve block system
    ● Cooperated with Thermaquil's hardware development team
  • Berkeley Pharma Tech 01/2021 - 05/2021

    Software Engineer Intern

    ● Implemented and deployed an ERC-20 token (digital currency) on Ethereum Network
    ● Built a full-stack web app to allow users to interact with the token
    ● Utilized React.js for the frontend, Node.js for the backend, and Solidity for the smart contract
  • UC Berkeley 08/2020 - present

    Research Assistant

    ● Trained CNN ML models to predict emotions using audio data from police body cams
    ● Developed models in PyTorch, trained on Microsoft Azure
    ● Achieved 87% cross-validation accuracy
  • Institute for Security and Technology 05/2020 - 11/2020

    Software Engineer Intern

    ● Trained NLP translation models for 3 languages and built a standalone APP for document processing and translating
    ● Used Google AutoML to fine-tune Google’s NMT models, Java for the standalone application
    ● Boosted translation accuracy on unseen data by 15-27% and document processing speed by 400%
  • Daily Chinese, LLC. 06/2018 - 03/2019

    Founder

    ● Directed a team of 6 including 2 Mandarin professors from Shanghai University to develop a Chinese learning App.
    ● Sold the app to Shanghai Yunxuan Information Technology Ltd., and is now the best-reviewed Chinese learning app on Apple App Store

Education

University of California, Berkeley

2019 - 2023

B.A. Computer Science
B.A. Cognitive Science

Highlighted Skills

I am currently looking for internship and research opportunities for the summer of 2022. Please feel free to contact me!

See my resume

My Portfolio

* Implemented an ERC-20 token that has the functionality of automatically burn, redistribute, and add to the liquidity pool * Built unit test framework * Used Solidity and Truffle for unit testing
ERC20 Deflationary Token

Solidity, Unit Testing

* This project aims to build a tool that could help doctors and patients to find medical resources that are informative for individual cases effectively.* This project was built using React (JS) Frontend and Flask (Python) Backend.

View Github page for more details.Chrome Web Store
Medical Diagnosis Helper

JS, Python, React, Flask, RESTful API

* 1 out of 15 people worldwide is suffering from a rare("orphan") disease. (For more information and our solutions, please checkout the github README page by clicking the github icon below.) * Built an open-source/enterprise tool that helps researchers/doctors/patients to analyze large sets (200k+) of papers. * Applied NLP and Graphic algorithms to visualize Biomedical papers and find potential relationships between automated annotated entities. For example, underlying links between DNA's.
Open-source Biomedical Literature Visualizer

NLP, Cloud Computing, Data Visualization, ML

* Implemented a 2D game with Java that randomly generates maps with seeds.* The goal is to find all the treasures within the map; main charactor has limited vision.

Check out Github repo for demo!
Explorer(2D-Game)

Java, Object Oriented Programming, GUI

* Designed and implemented a version control system that is similar to Git using Java.* Allows users to back up their files and maintains multiple versions of them, to collaborate with others locally, as well as remotely work with others over the internet.
Gitlet (Git version control systems)

Java, Data Structure

* Built a Chinese learning App (Daily Chinese) that had lesson plans designed by two of the most experienced Mandarin professors from Shanghai International Studies University that used speech recognition to assess learners‘ speaking abilities.* Lunched the APP within 3 months and received offers of $120,000 from potential investors. Determined to focus on school, later, sold the app to Shanghai Yunxuan Information Technology Ltd. for $50,000. * The App is now Super Chinese and is one of the most downloaded Chinese learning applications on the App Store.Super Chinese
Chinese Learning App (IOS)

Swift, App Development

* Trained a Convolutional Neural Network with 5000 images that had over 90% accuracy and built a web application for the model.Live Demo
Squirrel Species Classifier

Python, CNN, Image Processing

* Built the game in Java and implemented machine player with alpha-beta pruning and minimax algorithm.* Ranked 2/1700 in the tournament competing against other classmates’ algorithms.
Lines of Action (Game Tree AI)

Java, Minimax Algorithm

* Ever feel empty after talking to someone? Didn't know more about a person after talked to them? * Built an Web App with React that helps people to build strong connections with anyone.Live Demo
Big Talk Web App

React, Database

* Designed and implemented a soccer shooting game using Python's Pygame library.
Soccerrr(Pygame)

Python, GUI, Object Oriented Programming

Contact Me

Email: shelden@berkeley.edu

(331)-999-0314

Berkeley, CA