Cassie Huang

Actively seeking an entry level full-time job in software engineering. Experienced in fast pace environment and skill ready.
  • Skilled in JavaScript (ES6,Node.js, React.js), Python (Django, Flask), CSS and HTML in full-stack development.
  • Developed Full-Stack websites with frameworks such as Node.js (Express.js), React.js, Django, Flask etc.
  • Experienced with Docker, WebSocket, MongoDB, PostgreSQL, Redis, Kafka and Google Cloud Storage.
  • Deployed websites on Pass / IaaS such as AWS EC2, Heroku etc. Experienced with software testing.
  • Published Javascript package on NPM and Python package on PyPI. Experienced with microservice, CI/CD.
  • AI academic papers published by IEEE flagship journals, Gordon Scholar and Mary Millette Endowment Award recipient.

Projects

More About Me

Publications

1. Adaptive Quantization as a Device-algorithm Co-design Approach to Improve the Performance of In-memory Unsupervised Learning with SNNs,

S. Oh, Z. Huang, Y.Shi, D. Kuzum, IEEE Electron Device Letters, 2019.

2. The Impact of Resistance Drift of PCM Synaptic Devices on Artificial Neural Network Performance,

Y. Shi, Z. Huang, S. Oh, N. Kaslan, J. Song, D. Kuzum, IEEE Transactions on Electron Devices, 2019.

Contact Me

Please email me or call me according to my resume. I prefer email :) If you don't have my resume and still want to contact me you can open a issue here

and leave your email address & message. I will reply to you as soon as possible. Sorry for any inconvinence.

Copyright © Cassie Huang 2018

Tree Hole Community

  • Developed “Tree Hole”, a website where users are able to relieve stress and anxiety by sharing their concerns and getting advice from other people
  • Users can create/search profiles to find people with same background skills, and create/comment posts.
  • Implement backend in Node.js and MongoDB for fast prototyping and scalability.
  • Built web fronted in React.js (react + redux) for scalability.
  • Wrote unit tests with Jest, deployed on heroku to handle 1000~2000 search per second tested by Apache JMeter.

Close Project Visit Project

Calexy ChatApp

  • A chatapp where users can join chat room to chat with people in “many to many” or “one to one” way based on WebSocket.
  • Implement chatting recommendation based on birth date among users without disclosing any user information.
  • Built the backend with Node.js with the advantage of non-blocking I/O and lightweight
  • Used logic-less template mustache.js in frontend and wrote unit tests in Mocha for development productivity.
  • Deployed ElasticSearch on GCE to provide geo-location based search (e.g within 8 miles) and user info storage.

Close Project Visit Project

SharedRide Service

  • Developed a website which is easy and efficient for UCSD students to find and provide shared car rides.
  • Integrated Google Map API for location visualization and used autocomplete application for convenience.
  • Applied Google Cloud Storage for storing unstructured data (e.g. images) to take advantage of the Google cloud CDN service.
  • Built the backend with Flask for its light weight and large number of extensions.

Close Project Visit Project

Face Detection

  • Proposed and processed a new dataset for face detection and age/race/gender classification via pyTorch.
  • Customized deep learning models based on YOLO v3 and VGG16 to construct a two-stage deep learning architecture.
  • Achieved real-time detection and classification on a single GTX 1080 (on average 30 FPS).

Close Project

A Promising Movie

Successfully propose a cast & crew for next highest-grossing film. Please see attached slide for more information.

Close Project