Chengzhan Gao
高程展｜Graduate Student at UCSD
Software Engineer at Lyde
Hi! I currently work as software engineer at Ideker Lab, UCSD Health. In parallel, I am also pursuing Master’s degree in Electrical and Computer Engineering at University of California San Diego (UCSD). I bring a diverse academic background with a Bachelor’s in Civil Engineering from Tongji University in Shanghai, China. My foray into coding began during my undergraduate years, where interdisciplinary projects piqued my interest in the digital realm. The allure of creating worlds and defining their logic purely through code was irresistible, leading me to pivot towards Computer Engineering.
Throughout my career, I’ve collected a wealth of experience in software development. I completed internships as a software engineer at CRIB and Esurgi Biotech. I also collaborated with Professor Xie Pengtao on the LLM-related project DrugChat. My role as a software engineer at Lyde was particularly enriching, where I significantly contributed to the development of the AI-assisted university application tool, FoxiApply.
I am a stickler for perfection and detail. Beyond the technicalities, I firmly believe that software engineering is more than coding and I value communication and teamwork.
Interests
- Web Development
- Machine Learning
- Game Development
Education
MSc in Computer Engineering, 2024
University of California, San Diego
BSc in Civil Engineering, 2022
Tongji University
Minor in German, 2019
Tongji University
Skills
JS
80%
Python
90%
C/C++
40%
SQL DataBase
80%
React
80%
Java
40%
Work Experience
Software Engineer
Software Engineer
- Developed a 0 to 1 generative AI-powered platform to help students build their first resume for Ivy League college applications in less than 5 minutes without complex template writing, overseeing all phases from UI/UX design through to deployment.
- Optimized the performance of the website by integrating a Redis job queue to process multiple tasks at the same time resulting in a 12% reduction in job failures. Enhanced user experience and engagement by integrating i18n for multilingual support, resulting in 15% more user sign-ups from non-English-speaking regions.
- Applied Jest for unit testing and Cypress for end-to-end testing. Established a robust error-handling framework with CI/CD pipeline logging and alert triggers, minimizing critical SEV and post-launch bug reports by 24%.
- Employed Firebase for user authentication and data store. Integrated Stripe Payment seamlessly to the points system in the web application. Implemented Search Engine Optimization (SEO), improving the website impressions and clicks by 25%.
Software Engineer Intern
- Developed a data management platform for a health tech startup, enabling patients to efficiently manage exercise data from our proprietary Biostabilizer machine, with added data visualization for enhanced clarity.
- Optimized the app experience by implementing Redux for consistent state management and GraphQL for data retrieval, reducing server data fetches by an average of 50%.
- Enhanced user experience by cutting initial load times by 30% through improved front-end efficiency with Webpack and lazy loading. Additionally, build custom hooks to boost code reusability and abstraction.
Full Stack Engineer Intern
- Built a cross-platform App, CRIB, to foster roommate relationships through features such as real-time status-sharing, seamless expense splitting, and shared scheduling of house chores and responsibilities.
- Led the creation of key app pages using React Native, designed the user-friendly navigation system with React-Navigation, and established robust user authentication and database infrastructure powered by AWS Amplify.
Academic Experience
Graduate Researcher
Electrical and Computer Enigeering Department, University of California San Diego
- Led the fine-tuning of DrugChat, a GPT-like AI platform for pharmaceutical queries, using a novel dataset. Achieved notable enhancements in model accuracy and response quality.
- Executed a comprehensive data pipeline: extracted drug and protein information from the Wikipedia dataset, performed meticulous data cleaning and preprocessing, and prepared data for question-answer pair generation.
- Utilized GPT-4 for generating structured question-answer pairs from the processed data. Skillfully managed API limitations by implementing a throttle mechanism during data generation.
Projects
.js-id-featured
Movie Review Platform (Movieist)
Movieist is a demo application that displays user-friendly interface for movie lovers to leave their reviews and even watch the trailer to refresh their memories. This application utilizes React for the front-end and Spring Boot & MongoDB for the back-end and the database.
OFDM Communication System
OFDM is a mainstream PHY-layer technology widely adopted by modern wireless networks (e.g., WiFi and LTE). In this repo, an OFDM communication system in Matlab is implemented. The system represents a simplified version of the WiFi (802.11) PHY layer, which involves packet construction at the transmitter side, and packet detection, synchronization and decoding at the receiver side.
Revisiting and Refining AdaIN Style Transfer Model
Revisited the Adaptive Instance Normalization (AdaIN) style transfer, a significant advancement in the domain of style transfer. Introduced multi-level style transfer, applying the AdaIn module to various encoder levels instead of exclusively at the last level. Further enhancement was made by replacing the AdaIn module with an attention module, improving the model’s capacity to stylize images beyond color, also considering other features like texture and shape.
Analysis and Prediction of Graduate School Admissions
Every student applying for a college education surely wants to know what are the chances of getting into their dream school. Our project is designed to visualize and predict the probability of admission of certain students and the universities to which they apply, which is beneficial for the students in need.
Camp Compass
Created an Express framework-based website where people can share and fetch campgrounds information; Mainly employed AJAX technology, and RESTful APIs to handle HTTP requests. Provided the geolocation feature for users to view campgrounds by using Google Map API. Deployed a web service in Fly to make application scalable and maintainable.
Taichi Gravity Sketch
This project aims to take advantage of Taichi’s computational speed in simulating the physical motion of particles to implement a drawing game. Implement a game that allows players to manipulate particles and achieve some goals by themselves. Inspired by the idea of the game “Shadowmatic”, we decided to design the game by attracting particles to form a shape and scoring it against the target shape.
Netflix Clone
Developed a Netflix_Clone application on Xcode. Cloned the original Netflix App’s user interface and its basic functions. Used TMDB API to get information of TVs and Movies, and also used Youtube API to get access to videos.
Mini Computer Games Development (Unity)
Developed a spacing shooter game, where players drive a fighter in space and aim to shoot down as many enemy fighters as possible while avoiding being destroyed. Also developed a classic 2D grid-based side-scrolling roguelite game, where players adventure in the desolate underworld and they have to make use of resources to survive and fight monsters.
Machine vision based similarity modeling of rock fractures in tunnel face
Proposed a visualization method of discrete fracture network with randomly generated fracture trace maps based on machine vision methods and statistical principles using Python language, and thus established a complete discrete fracture network similarity evaluation system. Also proposed a pointing plus gridding algorithm to accurately calculate the position and density similarity of trace maps, innovatively took spacing similarity index into the calculation of comprehensive similarity. Finally, applied the established system to the actual project to test its reliability and applicability
CNN based Image Recognition for AsianGiant Hornets
Built an Auto-Regressive Integrated Moving Average model to predict the expansion of hornets; Trained a CNN-based model to distinguish Asian Giant Hornets from other bees; Implemented a multi-parameter optimization technique and finally F1-Score reached 0.992 while accuracy reached 0.987.
Earthquake Notification & Evacuation Guide
Built a database of earthquake occurrence and emergency shelters with web crawler technology using SQL, JavaScript and Python; Developed a software for earthquakes data visualization, real-time earthquake notification and emergency shelters router-guidance
Hornors
Shanghai Outstanding Graduate
This honor is given those students who have earned grades within the highest percentage of their school/department and achieved excellent academic performance during undergraduate study.
See certificateChina National Encouragement Scholarship
See certificateOutstanding Student of Tongji University
See certificateFirst-Grade Scholarship of Tongji University
See certificateAdvanced Individual
In College of Civil Engineering at Tongji University, only 10 students (undergraduate or graduate or PhD students) with outstanding performance in academic work or community work can receive this honor every year.
See certificateDean’s Award
This honor is to certify students’ academic performance. Only one student in College of Civil Engineering at Tongji University has the chance to receive this honor each year. One of prerequisite of this honor is to get grade A in all courses for three consecutive semesters.
See certificateChina National Scholarship
China National Scholarships are currently the highest honorable national scholarships available to college and university students, with the most standardized evaluation and the most stringent standards. Each year, more than 2,000 colleges and universities in the country, nearly 27 million students compete for 50,000 national scholarships, with only 0.2% of the awards.
See certificateChina National Scholarship
See certificateGallery
Contact
Leave me a messgae