I am a junior full-stack software engineer at Onehouse. I was previously at Goldman Sachs. I also built PageNow, a social Chrome extension, as a side project.
For the back-end, I improve performance by using techniques like multi-threading, asynchronous calls, and in-memory caching. For the front-end, I engage in user-centric design/development by analyzing user activities and acquiring direct feedback. For the infrastructure, I build a secure (strict access control and user authentication), scalable, and available system. I am also capable of diagnosing problems and resolving the issues.
- Working on the frontend and the control plane of the product.
- Developed a full-stack (Angular, Java, MongoDB) portfolio management tool managing $4B+ of assets.
- Led a project to build an order booking pipeline handling 250,000+ allocations per click. Used event-driven architecture with RxJava, Kafka, and Mongo Change Stream to update/present data in real-time.
- Served on-call duties during market hours to resolve bugs or issues within 30 minutes.
- Reduced the time to fetch transaction data from ~13s to ~2s (80% improvement) using multi-threading.
- Diagnosed high RAM usage and refactored services to be hosted on multiple servers with load balancing.
- Adopted an Angular test framework (ngneat) and created sample tests/patterns for the team to refer to.
- Built an open-source social Chrome extension for real-time activity sharing.
- Built a Python Fast API back-end to handle user information. Deployed the service (supporting load balancing/auto-scaling) to AWS with Terraform.
- Built a serverless API (AWS Lambda, ElastiCache (Redis), DynamoDB, API Gateway, etc.) for real-time features. Deployed with AWS CDK.
- Built the Chrome extension front-end with React and the PageNow client with Angular.
- Conducted surveys and analyzed user activity data to release features and UI/UX enhancements iteratively.
- Wrote technical blog posts on Presence API.
- Implemented group and domain level dotfile support and service port validation upon kernel creation.
- Developed storage proxy support for XFS and advanced features for the Backend.AI virtual folder system.
- Updated the shell script to achieve single-run installation and configuration to run Backend.AI.
- Head Teaching Assistant for IDS 157 (Statistical Inference) in Spring 2021-21
- Teaching Assistant for CS 121 (Relational Database) in Winter 2020-21
- Teaching Assistant for ACM 116 (Introduction to Probabilities Model) in Fall 2020-21
- Computer Science Classes - CS 122 (Database Implementation), CS 124 (Operating Systems), CS 142 (Distributed Computing), CS 144 (Networks)
- Machine Learning Classes - CS 179 (GPU Programming), CS 155 (Machine Learning & Data Mining), CS 165 (Foundations of Machine Learning and Statistical Inference)
- Applied Math Classes - ACM 104 (Applied Linear Algebra), ACM 116 (Probability Models), ACM 157 (Statistical Inference)
- AWS Serverless - API Gateway, Lambda, ElastiCache (Redis), RDS
- Python - Fast API
- Java - Jersey, Guice, RxJava
- Database - PostgreSQL
- Angular, React
- Chrome extension
- AWS - API Gateway, ElastiCache, ECS, RDS, DynamoDB
- Docker
- Open source code for PageNow.
- Composed of Presence API, Chat API, User API, chrome extension, PageNow client.
- Used Angular for the front-end and Python FAST API and AWS serverless stack for the back-end.
- A personal blog with posts about life and projects.
- Used Angular for the front-end and Google Firebase for the back-end.
- In 2022, ~3,100 visitors and ~9.7K page views.
- In 2021, ~600 visitors and ~2.7K page views.
- In 2020, ~100 visitors and ~700 page views.
- Developed a Chrome extension to promote serendipitous social connections by displaying web pages your friends are reading and by suggesting chats with anonymous users viewing the same web page.
- More details can be found at https://devpost.com/software/pagechat
- Developed a chrome extension that displays the number of COVID19 cases in the county that the user searches on Google Maps.
- Used React for the extension and AWS serverless (Lambda, API Gateway, DynamoDB) for the back-end.
- Developed a web app that evaluates the 'living score' of any coordinate in New York and provides a fair AirBnB rent price to guide users on finding good rooms at a reasonable price.
- Used ESRI's geometric feature layer aggregation with machine learning to calculate the score and price.
- Submission link - https://devpost.com/software/homes-reinvented
- Built a conveyor module that moves the center mass by a conveyor belt according to its slope.
- Wrote Arduino code that controls a motor according to the input from the IMU and IR sensor.
- Applied a complementary filter on the data obtained from the IMU to calculate the angle of the system.
- Improving Robustness of Reinforcement Learning for Power System Control with Adversarial Training - arxiv link