Gwen Au


Machine Learning Engineer


Profile

I have worked in IT and engineering companies which have inspired me to continue learning and implementing systems to advance automation. I aim to help design and refine advanced automation processes centred around machine learning and distributed computing to derive knowledge, optimise processes and enhance profitability. I like working with people, finding out what motivates internal and external stakeholders and applying creative and effective solutions. I look forward to contributing more to the tech community.

Experience

Career

Vicinity Centres

Machine Learning Engineer

November 2021 - Current


As a Machine Learning Engineer, I am responsible for leading the technical process design and development to deliver data science solutions to meet SLAs in an industry best practice manner. I deliver robust machine learning and AI solutions and services at scale and pace for effective use of these across the business. My role includes:
Process automation
Data architecture design
Data engineering
Data collection enablement
Data quality assessments
Data integration
Mentoring of other members of the data science team
Hands-on software engineering for full stack software development
My technical stack includes:
Python
SQL
NoSQL
Docker
Cloud technologies such as AWS, GCP and Azure

Data Engineer

September 2019 - October 2021


As a Data Engineer, I am responsible for Extracting, Transforming and Loading (ETL) complicated structured and un-structured data from disparate systems, into clean and easy to use data sets for whomever requires it. I work across departments, and assist the business with all their data needs, including: data modelling for business intelligence, supplying data sets for analytics & data science, and exposing API endpoints to customer facing applications and third parties. I use the cutting edge of digital technology to architect and implement system integrations via the API ecosystem and digital platform to set up Vicinity as a data-driven algorithmic business. Technical Stack:
Python and SQL for ETL processes in the Data Lake, Data Warehouse and Enterprise API environments
AWS and Azure Cloud Computing Services
Apigee API framework
Tableau for Data Visualisation


KPMG Austraila

Senior Consultant - Data Engineer

April 2018 - August 2019


As a Consultant for Data Engineering, my main responsibility was to engineer modern data solutions or pipelines to facilitate advanced analytics, visualisation & interactive data applications using the following tools.
Leveraging traditional and Big Data technologies on Azure and Google Cloud Platform to automate data pipelines from multi-structured and disparate data sources into timely and meaningful insights
Developing innovative web and mobile applications using tools including React, D3, JavaScript, NodeJS, Flask, speech recognition, HTML5 and CSS3
Solving challenging problems with advanced Data Science and Machine Learning techniques
Working with emerging and innovative digital solutions, including Artificial Intelligence, and Natural Language Processing


BushHQ

Full-stack Web Developer

May 2017 - August 2017


Created a progressive web app to locate and find out about local bush tucker with the following stack
Mobile development with the JavaScript Angular Ionic framework
Back-end development using Python, Flask, and PostgresSQL
Deployments on Heroku


Work Ninja

Full-stack Web Developer

November 2016 - May 2017


I worked as a full-stack web developer in an early stage startup for Work Ninja's software to manage the on-demand workforce. My responsiblities and technical stack included:
Engage in weekly agile sprints to help lead and implement all aspects of the software development
Make sure software is programmed to the highest possible standard to avoid disruptions in the live environment
Product management
Ruby and Ruby on Rails back-end
PostgreSQL Database
HTML5, CSS3, Sass, JavaScript, Angular 2 and AngularJS front-end
Mobile development with the Ionic framework
Git and GitHub for version controls
CodeClimate to validate code style, complexity, duplication and security
Test automation and Continuous Integration & Deployment (CI-CD) with RSpec and CircleCI
AWS webservers
Bugherd and Rollbar for bug and error monitoring
PivotalTracker for project management


JRNI (formerly BookingBug)

Deloitte Technology Fast 50 of 2015

UX and Front-end Developer

December 2015 - November 2016


Design, produce and deploy the user interfaces and technical solutions via HTML, CSS and JavaScript
Provide technical support and bug fixing

Endress+Hauser

Senior Project Coordinator

July 2010 - October 2013


Developed project processing procedures to effectively reduce instruction and clarification time between project managers and coordinators in order to prevent costly errors and delays
International project management including reporting, risk, opportunity and issue management

LEGO Group

Order Specialist

August 2008 - May 2010


Implemented a systematic order upload and batching to improve efficiency in order management, warehouse operations and minimise transportation costs
Represent clients’ concerns and opportunities at weekly sales and operation meetings
Manage the ordering, logistics and shipping process

Fresh Cane Pty Ltd

Operations Coordinator

June 2007 - July 2008


Developed a database system to record and forecast sales to help prevent spoilage and stock out
Inventory and freight management by streamlining delivery routes for lower transportation costs
Worked with operations staff to supply products to Sydney restaurants



Education

General Assembly

Data Science Immersive Course

October 2017 - February 2018


I have completed an intensive course on data science to improve the following skills in order to incorporate predictive modelling and advance automation for businesses. The skills can be broken down into these three main areas.

Data Engineering

Collect, extract, query, clean, and aggregate data for analysis
Identify big data problems and explore how distributed systems and parallel computing technologies are solving these challenges

Data Science

Build, implement, and evaluate data science problems using appropriate machine learning models and algorithms
Apply question, modeling, and validation problem-solving processes to datasets from various industries in order to provide insight into real-world problems and solutions

Data Analytics

Perform visual and statistical analysis on data using Python and its associated libraries and tools
Use appropriate data visualization to communicate findings
Create clear and reproducible reports to stakeholders


Web Development Immersive Course

June 2014 - September 2014


Web Development Immersive course. 12 weeks full-time course which, among other topics, covered Ruby, Ruby on Rails, Javascript, HTML5, CSS3, AngularJS and JQuery. My final project, Lumni, was created as a social media app to help graduates stay in touch by making it easier for them to plan events through communication tools such as chat, text messaging and links to other social media applications. They can also be updated with the latest industry news via an RSS feed, find out the weather and location through APIs and access information and photos from past and future events.


University of Western Sydney

Bachelor of Business Major in Marketing

2004 - 2007


Units Completed:

Introduction to Information Systems - Distinction
Strategic Marketing Management - Credit
Brand and Product Management - Credit
International Marketing - Credit
Analytical Reading and Writing - Credit
War and Peace - An Introduction to International Politics - Credit
Introductory Economics - Credit
Acounting Information for Managers - Credit
Fundamentals of Marketing Research - Credit
Marketing of Services - Credit
Marketing Planning Project - Credit
Consumer Behaviour
Marketing Principles
Management Foundations
Statistics for Business
Marketing Communications
Introduction to Business Law
Business to Business Marketing
Principles or E-Marketing
English, Text and Writing
Inventing Modernity
Foundations of Asia
Everyday Life
Literature, History and Culture
Media and Visual Cultures

Abilities

General

OOP Object Oriented Programming
Git Git for version control
TDD Test Driven Development
CI/CD Continuous Integration / Continuous Delivery
REST Representational State Transfer
API Application Programming Interface
ETL Extract, Transform, Load
Oauth Open Authorization
MVC Model-View-Controller Pattern

Design

Responsive Web Design
Prototyping with Adobe Xd
User Journey
Model Diagram
Site map
Wireframe

Project Management

JIRA
Trello
PivotalTracker
Bugherd

Languages

Python Python
Bash Born Again SHell
SQL Structured Query Language
JavaScript JavaScript
Ruby Ruby
HTML5 Hypertext Markup Language 5
Sass Syntactically Awesome Style Sheets
CSS3 Cascading Style Sheets 3

Data Engineering

Databases

Name Type
S3 Key–value
Redshift Relational
MongoDB Document-oriented
Azure SQL Database Relational
Azure Blob Binary Large Object Storage
Neo4j Graph Database Graph
MySQL Relational
PostgreSQL Relational
SQLite Relational

Cloud Computing

AWS Amazon Web Services Cloudformation, EC2, S3, Lambda, Data Pipeline, EMR, Cloudwatch, SNS, SQS, SES, API Gateway, Glue, IAM, Redshift, Route53, AWS Transfer
Azure Microsoft Azure Blob Storage, Function App, SQL Databases
GCP Google Cloud Platform Big Query, Google Compute Engine, Google App Engine
BD Big Data and Parallel Computing Spark, Hadoop

Data Science & Analysis

Scikit Learn Machine Learning
Jupyter Notebook Analysis
Scrapy Web Scrapping
Tableau Visualisation
Power BI Visualisation

Web Development

Back-end

Flask
Node.js
Ruby on Rails
Sinatra

Front-end

React
Angular
Ionic
Vue
JQuery
Bootstrap
Moment

Software Testing

Pytest
Selenium
RSpec

CI/CD

Buildkite
CircleCI
Travis

Other

Apigee
Heroku
Mandrill for emails
Twilio for sms

Programming Tools

Visual Studio Code
Command Line
SourceTree
Beyond Compare
Sublime

I always look forward to learning more...

Portfolio

General Assembly


2015 Top 50 Most Innovative Companies by Fast Company


Sentiment & Topic Analysis of Hostels in Vietnam

January 2018

Aim

Vietnam has become one of the most popular tourist destinations and businesses want to know what factors are important to ensure tourists to keep coming back to their hostel. Over 58,000 reviews were collected and stored in a PostgreSQL database. In a Juptyer Notebook, Machine Learning, Natural Language Processing (NLP) techniques (both supervised and unsupervised) were applied to analyse the data. Recommendations were created from these findings.

Methodology

1.Data collection through web scraping using a combination of Selenium, Scrapy spiders, BeautifulSoup, Regex, and PostgreSQL.

2.Sentiment Analysis using hostel ratings as the target. Packages such as Natural Language Toolkit (NLTK), SciKit Learn's CountVectorizer and Naive Bayes were used to analyse the sentiment. The model was validated through Train/Test Split and Cross Validation.

3.Topic Analysis using TfidfVectorizer and Latent Dirichlet Allocation (pyLDAvis).

4.Recommendations were made from the findings and conclusion.





Lumni App

September 2014


Lumni was created using Ruby, Ruby on Rails, AngularJS, JQuery, JavaScript, HTML5 and CSS3. I also made use of Google maps, Amazon S3, a temperature API and various gems (including Devise, CanCan, Oauth and Twilio). I created it as a social media app to help graduates stay in touch by making it easier for them to plan events through communication tools such as chat, text messaging and links to other social media applications. Users can also be updated with the latest industry news via a RSS feed, find out the weather and location through APIs and access information and photos from past and future events.




Wifind App

August 2014


The Wifind App was built using similar technologies to the above application. It was designed as a convenient tool for frequent travellers to access preferred wifi networks at their travelling locations. As it was a group project, my team and I made good use of trello, model diagram, site map, wireframes, and user journeys planning tools. We used git and github for version control and through great team work, we managed to create a responsive and marketable product in one week.




Word Shuffle

August 2014


Word Shuffle was a timed guessing game created in a 24hour hackathon. It was created with Ruby, Ruby on Rails, AngularJS, HTML5, CSS3 and deployed on Heroku within 24 hours. Try the game. See if you can guess the words within the time limit (hint: answers are related to web development). Good luck!



Interests

Dancing is my energetic creative outlet. I like the great outdoors for a little adventure but I am also delighted when I get to watch a piece of performing art.

Just for Fun

Hip Hop Dancing

2013


BookingBug Promo

2016



Entwined Bellydance Showcase

2019



Meerea Park Wine Promo

2020



Contact

Mobile:
+61 (0) 499 889 338
Email:
gwenau08@gmail.com
GitHub:
https://github.com/gwenau
Stack Overflow:
https://stackoverflow.com/users/4040580/gwen-au
Linkedin:
https://www.linkedin.com/in/gwenau
Twitter:
@Pock3t_Rock3t