Jun Alex Kagaya

Email: jun.a.kagaya@gmail.com
Mobile: (+64) 022 091 0069
Github: https://github.com/gieoon
Languages: Fluent in English, Japanese and Mandarin Chinese
Address: 34 Olympus Way, Richmond, Nelson

Education:

Graduate Diploma in Computer Science  2016 - 2017
AUT Auckland - GPA (9.0 scale): 6.25

Bachelor of Music – Major in Composition  2012 - 2014
University of Auckland

International Baccalaureate Diploma  Class of 2011
Suzhou Singapore International School, China


Work Experience:

Search Terms Manager - Flutter, Google Ads API, Flutter Redux
https://play.google.com/store/apps/details?id=webbi.ads.searchtermanalyzer

• Esoteric use case helping Marketers swiftly manage their Search Terms
• Compliance & Correspondence with Google Ads API team.
• Flutter for cross-platform development and deployment in Android + iOS

Freelance Software / Web Developer
Webbi Digital Studio, Nelson | August 2020 -
https://www.webbi.co.nz

• Delivered bespoke web designs/CMS to SMB’s in Nelson
• Collaborate/Communicate with clients to ensure their business needs are met by technology.

Automation Tester / Software Developer
Datacom Employer Services, Auckland | October 2018 - March 2020

• Increased the coverage of automation tests by 10% within 4 months,
• Ensured quality of features by working closely with developers and consultants.
• Constructed SQL queries / stored procedures for business analytics
• Improved existing analytics module’s accuracy by integrating custom trained machine learning

Web Developer / Tour Coordinator 
JTB Auckland | May 2017 - Dec 2017

• Managed client inquiries and communications in Japanese
• Developed and maintained inhouse projects.

Hololens Developer, Intern
Mobile Innovation, Datacom Auckland | Dec 2016 - Jan 2017        

• Was selected from over fifty students to build Hololens AR applications for the esteemed Microsoft Student Accelerator Program after being selected from 17 teams.
• Developed an interactive volcano experience for Auckland Museum visitors.
• Created a tourism application as part of a welcoming ceremony for Auckland’s Ngati Whatua Orakei Marae.
• Fortnightly reviews with clients and frequent showcases to internal teams

Projects:

Gyuck - React, Pytorch
https://www.gyuck.com

• Generate landing pages and blogs with AI.

NZ Commute - React, NodeJS, Leaflet, D3.js
https://nzcommute.com/
https://github.com/gieoon/NZ-STATS-Commuter-Competition

• Developed for There & Back Again Data Competition adjudicated by NZ Statistics

User Journey Precaching System - Tensorflow, LSTM, BERT, GPT-2, FastText, XGBoost, Gensim Word2Vec, Python

• Using a user’s prior sequence of actions to predict their next action
• Pulled millions of rows of raw user logs from SQL Database and processed it into actions to be interpreted by a model
• Improve application performance by forecasting and precaching expensive method calls.
• Requires real-time performance with minimal latency

Flight Risk Calculator - Python, Keras, Tensorflow, Scikit-learn, SQL, C#

• Developed an internal machine learning model with 70% accuracy in predicting the likelihood of an Employee’s termination in the coming six months.
• Supervised binary classification algorithm trained on over 7 million rows of SQL results for 470,000 Employees.

Covid19live.com - Dashboard tracking Covid19 and interesting stats
https://covid19live.com/

OhMyPerson.com - Dating site that uses no pictures, React
https://ohmyperson.com/

Slice & Dice Portions - Android (Java), XML, iOS (Swift)
https://play.google.com/apps/testing/io.github.juunoco.blobmobile

• Released a logical slicing game on Google Play Store
• 60+ unique & challenging levels
• Showcased to Game Development meetups with positive feedback.

J-Signal Web App & Mobile App (Android) - Javascript, CSS, JQuery, PHP, Google Maps & Geolocation API, Firebase

• Created a Web and Mobile App to monitor the location of Japanese tourists visiting NZ
• Realtime updates of hundreds of geolocation coordinates from mobile devices reported to the complementary web app at 5 minute intervals
• Achieved a 90% success rate during a 3 hour live demonstration to clients where 20 devices were distributed to 50 tourists
• Product garnered $3000 more in investment after generating strong consumer interest during Japanese tourism conferences.
  

More About Me:

• I have walked the Milford Track 26 times
• I play a lot of piano

References available upon request.