Seth Kim

Software Engineer

San Diego, California, United States4 yrs 5 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Proven track record in improving e-commerce user experiences.
Strong background in data analysis and visualization.
Passionate about teaching and inspiring others.

Stackforce AI infers this person is a Fullstack Software Engineer with a focus on E-commerce and Data Science.

Contact

sethykim@gmail.com LinkedIn

Skills

Core Skills

JavaPythonReact

Other Skills

AWSAWS AmplifyAirflowCCSSCognitoDDBHTMLJSONJavaFXJavaScriptJestKotlinLinuxMatplotlib

About

programmer, tennis coach. Very good at teaching and explaining things, love working with kids. I want to INSPIRE people Hobbies include: making music and learning about human anatomy physiology and sports kinesiology ig: @sethykim Todo: make my own youtube channel and website and share I have learned on there

Experience

Amazon

Software Engineer

Feb 2022 – Present · 4 yrs 1 mo · San Diego, California, United States

Improved recommendations experience for customers on amazon’s shopping site. Launched experiment that enables shoppers to see multiple widgets with titles “Most loved picks for you in X category”, diversifying recommendations, estimated $40 million revenue made (kotlin)
Created a data pipeline to smoothly facilitate flow from dataset upload to S3 bucket to DDB. Used MWAA Airflow to create DAG task workflows in python. (MWAA Airflow, AWS lambda, IAM, SQS, SNS)
Reduced latency from 140 ms to 80 ms when running multiple requests in parallel by implementing kotlin coroutines
To increase Amazon Prime signups, displayed ‘Join Prime’ button in different parts of the app. Estimated $2 million yearly revenue:
On android shopping app, built UI bottom sheet pop up using React Native.
On Desktop browser, added and styled ‘Join Prime’ buttons using HTML, CSS, Javascript, and Typescript
Built end to end integration tests and load tests. Reduced service fleet costs by 1/3 by transferring our load tests from TPSGenerator to Hydra platform (json, kotlin, typescript)
Incorporated bloom filter logic to reduce our team’s DDB batchGet costs and latency by 50%
Migrated our service to new region to facilitate lower latency and greater availability (Cloudauth, AAA, Route 53)
Reduced duplicative code by restructuring old legacy code to our new service format, enabling configurable json (kotlin)
Added dimensions such as “strategy”, “region”, “source” to our kotlin service’s metrics and created dashboards
To help marketers and engineers see useful widget info (location, links to the source), built Tampermonkey script

JavaKotlinPythonReact NativeHTMLCSS+10

Interpreta

Software Engineer

Nov 2021 – Jan 2022 · 2 mos · San Diego, California, United States

Frontend React

React

Deckard technologies

Front End Software Engineer

Jul 2021 – Nov 2021 · 4 mos · San Diego, California, United States

Refactored to multiple reusable components using ReactJS and React-Redux
Contributed to the timely production launch of Deckard’s rentals application
Helped make application fully configurable via JSON so that the UI dynamically changes according to each city's business rules
Led the testing by writing over 90% of the unit tests using Jest and React Testing Library
Tested formik validation for all configurable form components and navigation
Triggered resend verification code to email using AWS Amplify and Cognito authentication

ReactJSReact-ReduxJSONJestReact Testing LibraryAWS Amplify+2

Hcl technologies

Software Engineer Intern

Jan 2021 – Apr 2021 · 3 mos

Built Java Spring Boot API to enable stakeholders to 1) see a list of complete/incomplete tasks per sprint, and 2) predict percent of completed tasks for future sprints
Created a registration and login API for Spring Security authorization

JavaSpring BootSpring Security

Ucsd machine learning course: modeling and data analysis

Python Developer Data Science and Machine Learning

Jan 2021 – Mar 2021 · 2 mos

Formulated research question, tested hypothesis, analyzed the results in a meaningful and comprehensive way.
Performed data visualization using matplotlib scikit python libraries for linear regression. Generated and tested multivariate models with cross validation using an 80/20 split, created heat map to analyze relationship between independent variables and checked for multicollinearity
Tested which factors among 'transactionDate', 'houseAge', 'distanceMRT', 'numConvenienceStores' were the best predictors of housing price
Data Science Project- Examining exercise durations and average temperature per state in the US
In the United States during Spring 2020, did people exercise for longer durations in states with higher average temperatures? We answered this question by doing vigorous data cleaning, analysis, visualizations, and EDA using python libraries such as Numpy, Pandas, and Matplotlib to conclude that the lower the state average temperature is, the more likely an adult in that state meets federal guidelines for exercise.

PythonMatplotlibScikit-learnNumpyPandas