I made a community website for Koreans living in, or visiting, Las Vegas. These sort of 'community websites' are pretty common in the U.S. Korean communities. Discontent with the existing community site, I wanted to make my own.

My goal with the site was to digitally capture all information related to Las Vegas, for Korean locals and visitors alike. Information, such as things to do, things to see, where to go eat, list of sought-out businesses are all captured through the website.

The site is generating monthly revenue from sponsors who advertise through the website.

This is my first software project that actually made me money. Not a lot, but enough to get me into application development.

I got connected with a tour company that needed a custom content management system to manage their bookings and tours. They were using good-ole pen & paper, aided by Excel, and had trouble automating some of their processes. So I made a custom web application for them.

Little did I know then that when I would go on to start a web development agency later (Design Box), other tour businesses would approach me and ask for the same thing. And that's when Tour Booking App was born.

In electronics, a wafer is a thin slice of semiconductor, used for the fabrication of integrated circuits and, in photovoltaics, to manufacture solar cells. To manufacture wafers, engineers must follow a long, complex series of recipe steps.

I got connected to a process engineer who described the pain points of having to maintain the different recipes (and their variations), as well as having to keep work records for each recipe step. He and his co-workers were using Excel, but it was cumbersome to use Excel because it couldn't display images and videos when showing the steps, and also because sometimes they sometimes had a 'split batch' where they needed to mix different recipe steps from different recipe stages.

So I built this tool so wafer engineers can more easily manage wafer recipes and save their work logs.

This was a collaboration project with some friends. I set up the database and made the backend API, while others designed and wrote the frontend code using React.

We created a community website for students to post information like a shared bulletin board, so they can sell items, discuss courses, promote club activities, etc. with fellow students of the same school.

There are many ways beginner investors search for stocks to invest. Some invest in stocks of companies that they know or heard of. Some invest based on stocks' historical performances. Some follow advices of their friends, family, or advisers (on Youtube). Et cetera.

For me, I like to look at stocks' historical performance. But using the trading platform I was on, I could only search for stocks by supplying symbols or company names. I couldn't get a list of new, undiscovered stocks based on their historical performance criteria that I defined and supplied.

So I made this app so anyone can easily search and find stocks based on their own criteria that they can specify. It's a very basic app and it's missing a lot of functionalities, but I had a lot of fun making it.

This is a weekend project that I did for a social ice-breaker game that I used to play.

The game is called 'Liar Game' and the goal is to identify the imposter. To play the game, everyone is given an identical word except the imposter. No one knows who the imposter is, except for the moderator, who is whispering the chosen word to each participant. Each participant takes turn to provide a vague description about the given word to prove that he/she knows the word. The description must be clear enough to signal to others that he/she knows the word, but vague enough as to not give it away to the imposter who is actively trying to figure out the word. At the end of a round or two, if the participants cannot correctly identify the imposter, the imposter wins.

Unforutnately, this game requires a moderator who is not able to participate. So I built this app that randomly pick a word from a chosen category and randomly assign the imposter, so everyone can participate without the need for a moderator.

I struggle to express my thoughts and opinions in a large group settings. I remember sitting in a company meeting with more than 50 people, and the meeting organizer asked if anyone had any feedback on their new initiatives.

I had a few questions, along with some feedback that I wanted to share. But being timid, I held my tongue and just sat there quietly. I didn't want to appear stupid for asking stuff that everybody else knew.

That's when I thought it would be great if there was a way I could deliver my feedback to the management without the pressure of being judged or being put on a spotlight. This inspired me to create TruVoice, a tool that companies can utilize to receive anonymous feedback from their employees.

Many immigrants come to the U.S. to achieve their American Dream. However, filling out immigration forms is hard. While the forms are freely available from the USCIS website, the filing process is very confusing and difficult. So they often dole out thousands of dollars to lawyers to prepare their green card applications.

Having professional laywers is absolutely necessary when their cases require a legal advice. But often times, they just need assistance with form preparation and they are just looking for specific filing information that is customized to their immigration case.

So I created a software that help people file green card applications easily. Through a series of questions and pre-built decision trees, they are able to answer: "What forms and supporting documents do I need to submit for my immigration case?"

The idea is very similar to TurboTax, but it's for immigration filing.

I created a bot that grabs a list of contents to post on Reddit and programmatically post them to different subreddits based on pre-defined schedule. The bot grabs the list of contents such as the titles, body contents, subreddits to post, user accounts from a Google worksheet, and performs the content posting.

What's interesting about this project is that I can define the cron schedules for different posts in the Google worksheet, and the updated cron schedules will manifest during regular cron refresh cycles.

I imagine this bot can be useful for people who'd like to post recurring contents to different subreddits based on schedule (e.g. Redditors posting their services in freelance job subreddits, OF models posting their contents to grow their followers, etc.), without having to manually post the contents themselves.

In the past, I had purchased an Amazon Echo Dot on sale for $25. I liked the product, so I wanted to buy a few more, but the price had gone up to $50 per item. I wanted to buy them, but not at that price. I wanted to wait until there was another sale.

I requested a price drop alert for the product via CamelCamelCamel, but I never received the notification, even though the product went on sale a few weeks later. Why? Because CamelCamelCamel crawler wasn't tracking the price daily. It was checking intermittently (once per couple weeks) so it missed the 50% sale that I had been waiting for.

Hence, I decided to build my own. So I can perform price checks as frequently as I'd like.

I manage a large number of websites for my clients, and monitoring their websites is critical for their smooth business operations.

I created a script that can scan a list of websites to monitor their health. The script goes through a list of websites and gets a list of things to check for each website saved inside a Google Sheet worksheet. Once it performs the checks, it writes the results back to the worksheet.

Yes, I am aware that there are free services, like UptimeRobot, but I wanted to monitor more websites than the free plan allowed.

I sometimes get leads for website projects from Korean community websites, where people seek web designers/developers. I created a website crawler that scans different parts of these websites and look for certain keywords, such as 'web design', 'web development', etc. When these posts are found, the crawler sends a Slack message for notification.

I started this project while trying to marry three of my biggest passions: code, data, and basketball. I wanted to create a prediction algorithm that can reliably predict game outcomes and possibly even the betting spreads.

While in search for that “magical” algorithm, I performed various preliminary/explanatory analyses. In doing so, I found some interesting things, as shown by the data. I wanted to publish these findings and share them via a blog. And hence, Basketball 4 Nerds was born.

The General Social Survey (GSS) is a nationally representative survey of adults in the United States conducted since 1972. The GSS collects data on contemporary American society in order to monitor and explain trends in opinions, attitudes and behaviors.

To satiate my personal curiosity and practice data analysis skills, I performed various analyses and created visualizations to understand social trends and issues, such as how Americans feel about education spending over time, how their opinions regarding marijuana usage and religion changed over time, etc.

I noticed that many people in my community were looking for a career change into the world of programming. There are many free courses available on Youtube, but they all lack personal guidance from an expert. Bootcamps, on the other hand, cost $8K or more to go through. I wanted to create an affordable code school for people to pursue their learning for programming without breaking bank. Through a series of survey, I determined that they would be enough demand for an affordable programming school.

This was one of the projects that I was able to fail quickly. Ultimately, I decided to pull the plug because I realized that I'd be offering a service (not a product) and it wasn't ever going to scale. While I made no money, I learned many valuable lessons.