Sherpa

Sherpa provides last mile delivery services utilising a pool of drivers who can select jobs from the app on their own device.

I was the Senior Software Engineer for a squad comprising other Ruby on Rails engineers, mobile developers, QA testers and Product Manager. I provided mentoring to the junior enginers, worked with the product manager to translate business requirements into achievable pieces of work and with the wider technical group to deliver well crafted production ready code.

Sherpa operates a Ruby on Rails Monolith serving iOS and Android driver apps via API, ingesting customer orders by API, both directly and through middleware such as Shippit

• Ruby: 3.3, Rails: 7.1, Postgres
• Leadership: 2 x software engineers as direct reports, general team leadership requirements for the squad of 1 x Product Manager, 1 x QA tester,
• API controllers and serialization: Grape
• Internal administration: Active Admin
• Customer facing website views: ERB, Slim, Phlex, jQuery
• Async tasks: Sidekiq, Sidekiq Cron
• State Management: Acts As State Machine (AASM)
• Service Objects for consistent actions regardless of entry point
• Feature Toggling: Flipper
• Bug Tracking: Bugsnag
• Infrastructure: GCP
• Repo Hosting and CI/CD: Github
• Testing: RSpec
• Code Formatting: Rubocop

Mable

Mable are a 2 sided marketplace, connecting individuals with care needs (mostly NDIS or ACAT funded), with individual service providers. I was one of approx 40 Engineers working in Sydney, around Australia and also around the world.

The platform is a Rails Back End monolith serving an Angular Front End

• Ruby, Rails, Postgres
• Leadership: Provided mentoring to two junior engineers
• GraphQL: queries and mutations
• Infrastructure: GCP
• Repo Hosting and CI/CD: Github
• Testing: RSpec
• Code Formatting: Rubocop

Fat Zebra

Fat Zebra is a payment services provider, working with banks, acquirers, card providers and merchants. I was one of approx 15 Engineers working mostly in Sydney.

The platform is a collection of Rails monoliths (plus one Sinatra application) as well as several AWS Microservice applications. The principal task is to provide a fast response for a payment request

• Ruby, Rails, Sinatra, Postgres, AWS [Lambda, S3, SQS, Step Functions, SSM]
• Leadership: None. This was my first job as a professional developer. I was there to learn, and was so lucky to be surrounded by colleagues who were good at programming _and_ good at lifting the confidence and abilities of junior staff.
• Infrastructure: AWS
• Repo Hosting: Bitbucket
• CI/CD: Buildkite
• Testing: RSpec
• Code Formatting: Rubocop

Join The Dots

This personal project is an idea I had to collect, aggregate and analyse political donations, the people behind those transactions and any possible ways those people are connected.

It will always be a work in progress. Most recently I have added charting of donations and other transactions using Chartkick

• Ruby: 3.2, Rails: 7, Postgres
• Github Repo
• Live Site
• Hosting: AWS EC2 using Hatchbox for Deployment-as-a-service
• Domain Registration: AWS Route 53
• ETL: Data comes principally from the AEC transparencey register. It's cleaned with various custom scripts, and then imported using a rake task.
• Regex: Mostly at the point of importing of the data, regex are used to determine if a record is a person or a group, if the record represents a political party etc. For people we use regex to identify and remove titles and honorifics.
• Recursion: Crawling from person to connected groups and people, to their connected people and groups.
• Hotwire / Turboframes: Some of those recursive queries can take a long time, Turboframes are used to load slow data asynchronously
• Charting: Chartkick and Chart.JS
• Search: Full Text Search using pg_search
• Views: Phlex
• Testing: RSpec

Who The Bloody Hell Are They?

A friend had the idea to analyse the demographic data of the candidates for the 2022 Federal Election. The name is a play on the Tourism Australia ad "So where the bloody hell are you?" conceived while Scott Morrison was in charge of that body. We looked at data points such as Gender, Prior Career and Education. Volunteers tracked down the data, others looked after promotion (mostly Twitter). I built the website, with some help on the Front End

• Rails: 7, Ruby: 3.0, Postgres
• Github Repo
• Live Site
• Hosting: AWS EC2 using Hatchbox for Deployment-as-a-service
• Charting: Chartkick and Google Charts
• Search: Full Text Search using pg_search
• Other Interesting Gems: Gender Detector, HTTParty

Pitcher / Catcher

This is a pair of apps (in fact there are more catchers (1)) to demonstrate proof of concept for updating data in one app when data in another changes. An ex-colleague had the idea of a central library of technical documents, to which customers could subscribe, and their database (Catcher) would be updated automatically whenever a document was updated on the central library (Pitcher). Fundamentally the catcher subscribes to a webhook, that is sent each time the pitcher updates a document.

(1) I made a few different catcher apps, but basically they're all doing the same thing; accepting a webhook post when data changes on the pitcher. Not all catchers are subscribed to the same documents

• Ruby 3.0, Rails 7.0, Postgres
• Storage: AWS::S3 (with Ruby SDK Client)
• HTTP Client: Faraday
• Background Jobs: Delayed Job (uses Postgres for the queue, with a rake task to poll for jobs in the queue)
• Hotwire: uses Turbo / Broadcast to append or replace the html in the catcher app view when it is updated in the database (using Redis in the background)
• Github Repo (Pitcher) | Live Site (Pitcher)
• Github Repo (Catcher) | Live Site (Catcher)

Squash Scoring Demo

A re-creation of the mains coring page of this squash scoring app.

Really just documenting my React & Typescript learning journey. The app demonstrates the use of several React components, and the effective passing of state between them.

• Live Site
• Github Repo

Ruby on Rails on Raspberry Pi

Some basic sensor and hardware interaction with Ruby on Rails on a Pi.

interactive use, by visiting /led/on and /led/off, simply turns on or off all outputs

automated, there is a crontab entry that fires a rake task every minute (`/lib/tasks/breadsched.rake`). The rake task reads the value of the temp sensor and sends a request to /heater/on or /heater/off depending on the value.

webhooks; listens for incoming requests on /hook/github. I configured the github repo to send a webhook post request for any event on this repo. I used ngrok to allow incoming requests to my server.

• Ruby 2.7, Rails 7.0, SQLite3, ngrok
• Raspbian Linux
• Github Repo
• Slide Deck

Picture Liker

This is a demo project for a design company who were interested in capturing likes and dislikes of their clients prior to commencing work.

AdminUsers have full CRUD access, Users can like or dislike a number of pictures.

• Ruby: 3.2, Rails: 7, SQLite3, Sidekiq
• Github Repo
• Live Site (visit /users/sign_up to create an account)
• Hosting: AWS EC2 using Hatchbox for Deployment-as-a-service
• Domain Registration: AWS Route 53
• Authentication: Devise
• Administration: ActiveAdmin
• File Storage: ActiveStorage (AWS S3)
• CSS: Tailwind
• Email: Sendgrid
• Hotwire / Turbostream: When a picture is liked/disliked, Turbostream is used to update the status of the clicked button, and also to disable other buttons once the limit has been reached
• Image Recognition: AWS Rekognition. Using the AWS SDK, the image is sent to Rekognition, and the resulting labels saved back on to the picture record.

Sonos Blame

Answers the question: Who put that terrible song on the office sonos?, well it does if all team members are signed up to this app...

Login to this app with your spotify login, and it will pull all your listening data. You can then query the app with an artist name, and it will return the user who is the biggest fan of that artist. Obviously not a very serious project, it does offer some interesting insights into your Spotify tastes.

• Ruby: 2.6.4, Rails: 6.0, Postgres, Redis, Sidekiq, OAuth, RSpec, Faker
• Github Repo
• Live Site: Currently offline

Nature Color Schemes

Some fun with colors; I got the color values from some photos I had taken and had some practice with CSS variables and also trying to make some complementary themes.

• HTML, CSS, Bootstrap
• Github Repo
• Live Site on Github Pages

divBow

Some fun with maths, colors, jQuery and time based animations.

• HTML, CSS, jQuery, CSS Grid
• Github Repo
• Live Site on Github Pages

RSpec Tutorial / Notes

My RSpec notes, which I started when I was a junior, have added to over time, and which I have provided to several junior engineers over the years now

• RSpec Document

DataHouse & Pipeline

Before I learnt any Ruby or Javascript, I created and sold 2 Software Products; DataHouse, an Inventory Management system including Quotes, Invoices, Purchase Orders and Pipeline, a Recruitment Management system

Both of these products were built in Microsoft Access, tightly copupled with MS Outlook for Email and Calendar functions.