About 🙋🏻‍♂️

Hi! I'm a third-year student at the University of Washington, Seattle, studying both Informatics (Software) and Business (Operations Supply Chain). I also work at UW Research Compliance as a data analyst and am the current Director of PR for UW IFC.

I'm highly experienced in research, data analysis/collection, marketing (graphic design, social media and promotional videos), full stack software development, database management, ML and AI et cetera. I'm also experienced in DJing.

This summer I will be working at Amazon Bellevue

The best way to contact me is through my email: mattcng9 (at) uw dot edu

Fast Facts

• Computer Languages: Python, Java, JavaScript, C++, Bash, TypeScript, HTML, SQL, R, JSX, SCSS, LaTeX, Markdown
• Human Languages: English, Vietnamese, Mandarin Chinese, Spanish
• Tools: Git, Pandas, Seaborn, plotly, scikit-learn, React, FastAPI, Next.JS, Node.JS, PyTorch, Jupyter, Agile framework, QuantConnect, MySQL, SQLite, HuggingFace
• Software: Davinci Resolve, Figma, Canva, Adobe Photoshop, Tableau, Excel, Microsoft Office Suite
• Interests: Data science, fullstack development, statistics, quantitative finance, fintech, defense applications, graphic design.
  EDM, College Football, weightlifting, game theory, Taekwondo, cooking.
• SAT: 1590

I am open to learning any other (computer or otherwise) languages requested.

Courses I've taken at UW include Data Structures and Algorithms, Operations Management, Artificial Intelligence, Product Design, Product Management, Data & Database Modeling, and Information Systems.

Projects 💻

I've always been interested in quantitative finance, so this project allowed me to explore that intersection of CS and business. Initially I trained a model through the PushShift torrents, which let me compare historical market data to sentiment (measured by HuggingFace) on Reddit that day for a particular ticker. Eventually, because my model was longer-term (i.e. not updating every second), I switched from a Selenium/BeautifulSoup hybrid collection model to the pushpull.io API directly. I've backtested the algorithm (with historical data) and it beat the market by 39% yearly, but I'm aware that a live version won't produce as-stellar results.

I wanted to generate snowflakes for an IFC post, and also had additional winter-themed posts in mind for other organizations. I couldn't get the snowflakes to sit right manually, so I looked into the Figma API and found that it would be possible to make a snowflake generator plugin. The result was a plugin that used Perlin noise to generate a pattern of snowflakes, customizable by number of objects, size, and spread.

Used fastAPI and python Requests to pull the top 100 reddit posts with a specified keyword. I implemented a responsive frontend with React and deployed with Next.JS and Vercel to ensure maximum stability and uptime for the live website.

My team and I trained three machine learning algorithms, k-NN, Random Forest and Logistic Regression to predict whether or not an individual had diabetes based on a variety of biometric factors. I also created a web-app for an interactive kNN plot that allows users to adjust the "k" as well as overall sample size.

Academic Projects

Pac-Man ML

Designed + implemented ML concepts in the context of a Pac-Man game:

• Search algorithms like BFS, DFS, cost function and A* search
• Minimax, expectimax, reflex, and alpha-beta pruning agents
• Value iteration, policies, Q-learning and Epsilon-greedy approaches
• Perceptron classification and feature design

Food Insecurity: Project Sigma

A personal interest of mine has been food insecurity and its effects on the homelessness cycle. My team and I designed and implemented a MySQL database that contained information on clients, dietary needs, nearby resources that provided those foods and their prices, and the nutritional information of those foods. We also created useful functions to aggregate commonly accessed data and help nutritionists best meet their clients' needs.

We prepared a full schema and presentation alongside our actual server-side SQL code.

NutriDash

Stemming out of the general food desert problem, NutriDash was aimed at creating an accessible app for elderly users who may not have the resources to accurately inform their unique food needs. After tedious research to accurately determine user needs and goals, we prototyped our fully interactive app in Figma and prepared an accompanying presentation.

Registration System Redesign

Given a scenario, my team and I used the Agile Methodology to fully lay out the development process behind redesigning a university's outdated registration system. We:

• Identified stakeholder (Role and User) needs
• Wrote comprehensive user stories, functions and epics
• Created architecture requirements based on needs and current standards
• Determined objectives and key results and identified the risks and assumptions we made
• Designed UI, Dataflow and workflow diagrams

Experience 📝

You can access my full resume by emailing me

This is also the best way to get more information on my current positions

Current

UW Research Compliance

Data Analyst

• Wrote Python scripts to conduct weekly analyses of team performance datasets, generating updates pushed to a public site
• Grew invoicing output by publishing reports based on data from central SQL DB.
• Utilized Tableau to highlight trends and improve AR team turnaround.
• Automated folder extraction/filewriting, increasing efficiency with Batch scripts

Montlake Consulting Group

Consultant

• Use data to gain insights on niche markets for clients

Pi Kappa Alpha Fraternity

Head of Public Relations

Former

Husky Taekwondo

Head of Social Media

• Grew Instagram following from 600 to 900+ in 3 months of tenure
• Directed, edited and shot a recruitment video reaching over 10k accounts and increasing club sign-ups by 80%