Building a peer-to-peer English learning platform - learners hold each other accountable and learn effectively at a much lower cost. Raised $1.2M Pre-seed round. Backed by Greylock Ventures, Reach Capital, Lombardstreet Ventures, Matterscale Ventures, and Techstars Ventures.Helped learners from 140 countries around the world to achieve confidence & fluency in English.Techstars 21'www.immigo.io

-Created a REST API that generates permutation of an image given PNG files containing parts of the image with different texture utilizing Flask. -Implemented an auto image cropping script through Canny edge detection utilizing Sobel filter.-Developed a multifunctional image editing tool using KonvaJS and HTML Canvas.-Implemented a login validation for React Native mobile application-Implemented an optimized search functionality in the mobile application, through leveraging debounce.

-Deployed company-wide collaboration platform where traders and engineers can comment, like and favorite proposed features from Redmine leveraging React.js, and Flask.-Implemented a faceted search algorithm to allow users to search and filter by specific feature categories.-Implemented dynamic keyword highlighting using Regex matching.-Implemented a lightweight store to maintain global states, and separate data handling logics Belvedere Trading is a leading proprietary trading firm located in Chicago, specializing in equity index and commodity derivatives.

-Developed a transaction platform to streamline the bill splitting process between a group of users by building and integrating an intelligent algorithm that settles group expenses based on accumulated transaction activities within a group.-Created a REST API leveraging Node.js and MongoDB, to manage platform’s user database and various transaction information.-Worked in Agile environment, with two weeks long Sprints iteratively.-Created a WebSocket server to allow real time user interaction.-Pushed containerized application to docker registry in Kubernetes. -Integrated a Redis Pub/Sub server to handle scaling WebSocket servers

-Used data scraped from market intelligence data feeds and Python algorithm to determine high net-worth individuals for clients seeking investors -Built user-interface through React displaying heat map, data table, and tree map displaying transaction and potential investors' data stored in SQL database

Taught English, math, chemistry, and physics to Korean high school students

-Wrote an algorithm in C++ to automate the calculation of protein quantities within various samples in order to determine the correlation between activities within hypothalamus and insulin production -Presented research results during 3 months’ period with other colleagues, to other researchers/professors.