Personal Projects
I am someone who enjoys staying busy and my preferred way of doing that is by working on projects. I start many, I finish some, but I always learn something new — which is kinda the point.
User Input Parser Python Package
While contributing to the Blankly project on Github, it was suggested that I make one of my contributions a python package. The code I wrote parses the settings values set by the user to ensure they conform to the developer’s expectations and if it doesn’t confirm, a helpful error message is printed. This was my first time authoring a python project, auto documenting code, and my first dive into Continuous Integration / Continuous Deployment.
You can checkout the project at the Github Repo.
Dynamic Portfolio Allocation
I have been fascinated by the idea of algorithmic trading for a while and set off to build my own trading engine quite a few times; then I stumbled upon the python package Blankly. Blankly provides a treading and back testing engine that abstracts the broker so you can trade cryptocurrency or stock market assets. I dove into the source code and built a portfolio allocation framework on top of it so i can define and test multiple different trading strategies and then allocation capital across them depending on their historical performance.
Stock Screening Dashboard
Wanting to learn a little more about technical analysis of stocks and Python, I set off to build Algorithmic Trading capability that would execute trades based off user-provided indicators. Currently, I have a dashboard that shows the OHLC data for a specified company along with the associated technical metrics. The next steps are to integrate sentiment analysis and machine learning to predict the future value of the stock and then implement automatic investment / divestment (with fake money first)
The dashboard is written in Python and was my first time using Dash and Plotly.
Autonomous Tank
The objective of the project is to create a Simultaneous Localization and Tracking (SLAM) robot that can map the environment with an ultrasonic range finder. This was the project i took on to teach myself how to interface with hardware from the ground up; reading an IMU’s documentation and writing software to community with it over SPI. It was also my first experience with multithreading in C++ and implementing an Unscented Kalman Filter (UKF) within the navigation algorithms.
MagicMirror
Made from an old computer monitor, a RaspberryPi and some custom electronics, this project dragged my old bedroom mirror into the 21st century. The MagicMirror is a 2-way mirror that has a LCD screen behind which can be seen from the front side of the mirror. This was the project I took on to learn a little about running software on a RaspberryPi.
Other Notable Projects
Stock Market Web Scraper
One of my very first introductions to software was writing a web scraper to pull stock market data from a website. I pulled a couple of key parameters, based off a book I was reading at the time, and checked if the stocks met that condition. This was my first experience with web scraping and the software worked as expected, but I didn’t dare put any money behind my predictions.
Smart Lock
Wouldn’t it be great if your front door just opened when you walked up to it? That was the goal I was trying to achieve with the project. I 3D printed a few gears and a housing, as well as setup wireless communication between devices to make it happen. This project taught me how to setup a TCP/IP Server-Client communication scheme.
3D Printer
“Man, if I had a 3D printer, I could make so much stuff!”… or so I thought. Turns out buying a cheep 3D printer and upgrading the firmware does not make it work any better that it did before; just adds a few features. I was able to flash the firmware no the printer with some open source software and pair that with a RaspberryPI. The end result was wireless printing and touch-screen control of the printer.