Architect and developed new features for a large-scale SaaS.
Architect a Multi-tenant SaaS application on AWS. Created a code base with Clean Architecture in mind with a light-weight Domain-Driven Design architecture. Encouraging colleagues to write unit/integration tests and integrated Codecov with GitHub PRs to motivate them by visualizing coverage percentage. Also created the code base on how to write e2e tests with Cypress for the front-end engineers. In addition, created the pipeline for CI/CD with CircleCI in combination with Cloud Formation. The pipeline also creates a Review App on every Pull Request to make the review process as smooth as possible. Finally, developed an authorization service using Open Policy Agent as a Go library which has a feature similar to the AWS IAM and also developed a Lambda Authorizer in order to authorize multi-tenant requests in the API Gateway.
Created a web application to communicate the concept of an in-vehicle device system for a major automobile manufacturer. Created using a combination of Angular and D3.js to build a uniquely designed graph.
Planned and tested solutions to migrate a B2C service from a Japanese cloud provider to AWS. Including a NAS migration from a single LVM node to a DRBD cluster with 2 nodes using LVM. In addition, converting provisioning shell scripts to Ansible and automating AWS resource deployment with Terraform. While it took days to prepare a full environment, using Terraform and Ansible shortened it to hours.
Led a team in creating an Online Learning Management System for children to learn electronics. All processes used containers and the system was deployed using Rancher (Cattle as the orchestrator and RancherOS for the OS). Set up CI/CD using GitLab and GitLab CI, ran tests using Selenium Grid. Developed the frontend Single Page Application with Angular and the backend using Node.js.
Developed a Spark application for a travel time prediction system. The machine learning logic was highly customized using R and our team converted it to a Scala Spark application. While the R version took minutes with large amounts of data, we were able to reduce it to around 30 seconds by taking advantage of scaling out the spark executors.
Led multiple projects using Marionette.js and AngularJS. Also built iOS and Android applications using Ionic.
Created a travel route analyzing system as a back-end developer.