It’s been a long time since writing the initial bio and things have changed a lot so here is what I’ve been up to lately.
Education
While the goal was to aim for Data Science, and I began a Bachelor of Computer Science to do just that, several projects at work as well as the coursework quickly taught me that my heart was closer to Data Analytics. Due to the slow speed at which the course was going while doing part-time and the number of learnings from both personal and work projects, I decided to exit with a Diploma of Science and Technology and enrol in a Graduate Certificate in Professional Data Analytics at Swinburne which I competed in 2022 and now completing a Masters of Analytics with UNSW to be completed later this year.
Professional Projects
Python & SQL
During my time at View Hotels, there were several software updates to move to cloud-based software which allowed us to gain access to more data than ever before. While in lockdown I built a data pipeline taking from the Property Management System, Restaurant Point of Sale and VOIP Phones in order to compile the data into a central database.
As the data for these systems is very fragmented and not from the same software supplier, a lot of data cleansing and processing had to be done in order to bring it together in a fashion that was useful.
The Property Management System Infor HMS has APIs available however they have limitations and the data types can change without notice, this taught me a lot about continuous integration and data validation.
The Restaurant Point of Sale provided a nice API however the data was very fragmented, for simplicity of data pipelines I chose to warehouse it somewhat as is apart from replacing strings with IDs. Later, this was replaced with a new POS system which unfortunately did not provide an API, so the data had to be emailed each morning, then cleaning the data before warehousing.
The VOIP Phones provided an API however it was very unreliable, this once again required learning a lot of continuous integration so that even if there were disruptions with the API, reattempts would be made and the missing data would be handled gracefully.
Power BI
Several options were explored and prototypes were made in Tableau, Google Looker, AWS QuickSight, Yellowfin BI and Power BI. Finally, Power BI was decided due to the flexibilities of distribution, the elegance of handling desktop and mobile and the ease of development.
While the data was cleansed during the capture process, Power BI’s power of doing extra preparation of data types, joining tables and other manipulation via Power Query, then the flexibility of doing the calculations via DAX allowed for building customised dashboards for all departments which were updated on an hourly basis.
Software Integrations
During my time with View Hotels, I had the opportunity to lead the team in integrating a new Restaurant Point of Sale – Shiji Infrasys, a new Events and Sales Management Software in Infor SCS, a new key system with VingCard built Melbourne’s First Gaming Room. The project management was challenging at times with many of these projects intersecting but it was a great experience and taught me a lot.
Excel
While I believe there are often better tools available for longer-running projects, sometimes Excel is all that you have available. Over the years I have done several projects ranging from web scraping and filling in online forms thanks to VBA macros, data integration cleaning from SharePoint via Office Scripts to creating attractive interactive dashboards.
Personal Projects
Regatta Racer
I had never contemplated building an app for either mobile or watch until I bought a new smartwatch and started to use it for sailing. Race starts are busy times on a sailing boat, races can be won and lost by either getting a good or poor start. It is also invaluable to navigate the course accurately. I found that there were no good apps on offer for the watch, so decided to investigate the possibility of making one. The operating system is Tizen and I decided to program a web app in a combination of HTML and JavaScript. I quickly had a prototype up and running and quickly realised that I could refine it for distribution which became Regatta Racer.
It’s now been up for sale since December 2019 with a few updates and downloads still flowing through.
The newer watches are now on Wear OS. I initially developed a version in Kotlin but I’ve recently been developing a new version of the app for these watches, I chose Flutter this time as the language because it allows for the development on phones and watches for both Apple and Android simultaneously. The updated release should be going out in the next few weeks.
Sailing Flags
While I was in the process of building Regatta Racer I was doing research on distribution on the Samsung Store and realised there were likely to be a few kinks to work out so it would be easier to try with a simpler app.
I decided to do a simple Sailing Flash Cards app, and while sitting in the airport to fly home decided to challenge myself to complete an app while waiting for the plane and the flight.
It was a simple game involving scrolling to select the flag and letter combination and a definition of the International Maritime meaning as well as Regatta Signals.
I came close to completing it during the flight with a bit of refinement when I got home and distributed it as my test project.
Later I developed a flashcard app for the phone to help memorise the flags, their corresponding letters and meanings.
Current Projects
Simplify Data
Since leaving View Hotels I have embarked on running my own company, Simplify Data supporting businesses with their data and automation needs.
One ongoing client has been a winery, Marnong Estate where I’m working with them on cost control for the F&B teams as well as their reporting needs.
I am still providing ongoing support for View Hotels for their reporting suite as requirements change and there are updates to the provider’s software requiring updates.
I’m also working on a custom BI dashboard in Python for hotels, there is a lack of visibility in data in the sector and those that are providing tools for data visibility are unaffordable for smaller operations.
Leave a Reply