Liam Fayle

Jan 9, 2024

Personal Website liamfayle.com

My personal website is built using Node.js, Next.js/React, providing users with a dynamic and responsive experience. It utilizes a NoSQL database, specifically MongoDB, to store my projects as objects. Leveraging Docker for deployment and management ensures efficient scaling and maintenance. This setup streamlines development workflows, enhances reliability and scalability, and enables seamless updates and optimizations.

Dec 18, 2023

Swiftbooks Bookstore

SwiftBooks is a comprehensive web application offering a virtual bookstore experience. Users can explore and purchase books with ease, thanks to features like a shopping cart and a streamlined checkout page. Personalization is key, as users can create and manage their own book lists, complete with reviews. Admins have control over the platform, with the ability to manage accounts and reviews. Sign-in options include both traditional username/password and Google account integration. The front-end, built with Next.js, React, and TypeScript, offers a seamless user experience with the Mantine UI library. On the backend, Node.js and Express power the application, with PostgreSQL handling data storage. The communication between front-end and back-end occurs via a REST API. The project is deployed using Docker, with separate containers for the front-end and back-end, ensuring scalability and efficiency. Testing ensures reliability through unit and integration tests.

Dec 5, 2023

ASP.NET Excel Clone

Simple Excel clone using Blazor, ASP.NET, and C# within the .NET Framework. This web application emulates core Excel functionalities, enabling users to fill cells and perform basic arithmetic operations. Users can input numbers directly into cells, reference cells in calculations, and utilize built-in functions for summing and averaging numeric ranges. This tool provides a straightforward, interactive interface for managing and computing data directly in the browser.

image of train test accuracy over epochs

Nov 28, 2023

Python

Keras

Tensorflow

CNN Classification of Brain Tumor MRI Images

Investigates the use of Convolutional Neural Networks (CNN) for classifying brain tumors in MRI scans. Developed a CNN model that classifies tumors using grayscale 2-dimensional MRI images, achieving a notable accuracy of 90.84%. This achievement is particularly significant as it was accomplished without the use of transfer learning or 3-dimensional data, leading to a much leaner model. The study employed data augmentation techniques to enhance model robustness and accuracy. Despite minor imbalances in the data categories, the model proved effective across various tumor types. The work highlights the potential of simple, efficient CNN architectures in medical imaging, offering a viable tool for aiding diagnosis in resource-constrained environments.

Mar 27, 2023

Python

scikit-learn

PostgreSQL

Counter-Strike Automated Betting System

The CS:GO Automated Betting System is a project designed to optimize betting strategies for Counter-Strike matches. It employs a custom-built web scraping API with anti-bot detection to gather match data from various websites, storing it in PostgreSQL data tables. Through data extraction, models are constructed and refined to predict odds more accurately than those offered by sportsbooks. A custom Selenium web spider continuously monitors live odds on pinnacle.com, seizing opportunities by betting on the earliest and most favorable lines available. To overcome bot protection measures, 2Captcha is integrated into the system.

Mar 2, 2023

Python

PostgreSQL

Option Data REST API

This project is a comprehensive system developed using FastAPI, aimed at storing and managing over 2TB of daily options data, including contract and price information. It features an extensive API that supports various endpoints for accessing historical data related to options trading. These endpoints include fetching historical expiries, strikes, prices, quotes, implied volatility (IV) rank, volatility cones, earnings, and IV information.

image of homepage of the application to control rooms

Feb 8, 2023

Smart-Home Individual Room Temperature Control Prototype

This project utilized small microcontrollers connected to motors on individual room vents to regulate airflow. A central control module acted as a web server, receiving data from the microcontrollers over WiFi. Using a web app, users could set desired room temperatures, and the central module adjusted the vents accordingly to maintain comfort levels.

Jan 10, 2023

VHDL

MIPS 16-bit Processor Architecture

Constructed a 16-bit MIPS CPU using VHDL began with individual gates and progressed to more complex components like multiplexers, adders, and registers. Modules such as the ALU, control unit, and sign extender were developed incrementally. Files like ALU_16Bit.vhd, Control.vhd, and MUX31.vhd were added to reflect each module's creation. Testing and debugging were integral, ensuring functionality at every stage. The project culminated in a fully functional CPU using the MIPS ISA.

Image showing the hedgebands resulting around potential stock prices.

Dec 11, 2022

Python

Zakamouline Optimal Delta Hedging Under Transaction Cost Constraints

This project implements Valeri I. Zakamouline's Optimal Hedging of Options with Transaction Costs method. It utilizes a Black-Scholes-Merton model for pricing options. Users can obtain hedgeband values at the current spot price and for a range of spot prices. The hedgebands are calculated considering proportional transaction costs and a risk aversion parameter. Interpretation involves adjusting the position delta by buying or selling shares to maintain it within the specified hedgebands, minimizing transaction costs. References include Zakamouline's work and Euan Sinclair's "Volatility Trading: 2nd Edition."

Graph showing growth rate G on Y axis with increasing bet size f on X axis. A red dot represents the vertex of this curve and optimal bet size f.

Dec 3, 2022

Python

Kelly Position Sizing for Options Using Geometric Brownian Motion and Monte Carlo Simulation

This project implements the Kelly Criterion for options positions, aiming to maximize the expected value of log wealth. It involves generating geometric Brownian motion paths, calculating terminal spot prices, and determining the return on premium at expiration for each path. By assessing various bet sizes, it identifies the optimal bet size (f*) that maximizes the growth rate. Results include growth rate curves, sampling distributions of optimal bet sizes, and mean returns at different forecasted volatilities, providing insights into position sizing strategies for options trading. References include Edward O. Thorp's work on the Kelly Criterion.

Image of the black scholes merton equation with descriptive text pointing to its components.

Nov 27, 2022

Python

Black Scholes Merton Model - Python Implementation

The Python code implements the Black-Scholes-Merton (BSM) option pricing model, allowing for flexible pricing and analysis of option contracts with customizable parameters. It calculates option prices and Greek measures, supports both long and short positions, and employs the Newton-Raphson method for volatility approximation. Advantages include flexibility, comprehensiveness, readability, and ease of use. However, it lacks support for American options and may require further optimization for numerical stability. Integration with real-world data sources would enhance its practical applicability.

Nov 12, 2022

Assembly

FreeRTOS Real-Time Security System

This Real-Time Security System project utilized an STM32F429I-discovery microcontroller along with a PIR motion sensor and an Arducam Mini Module Camera Shield. Developed in C code, the project interfaced these components to display video on-board upon detecting motion. The PIR motion sensor was connected to GPIO (PC8), while the Arducam utilized SPI (SCK, CS, MISO, MOSI) and I2C (SDA, SCL) interfaces. This setup enabled seamless integration and operation of the security system.

shows a simple graphic of bitcoin book depth chart.

Jan 12, 2022

Python

Automated Bitcoin Exchange Bot Using Custom Interac E-transfer API

This project is a Python-based Bitcoin bot leveraging Selenium for web automation, 2Captcha for captcha bypassing, and stealth techniques like IP and user-agent rotation. It operates on LocalBitcoins.com (where the premium / spread for transacting in interac etransfer was significantly wider than on large dollar denoted exchanges), profiting from the spread between buy and sell prices, typically earning 5-10%. To mitigate inventory holding risks, it hedges by purchasing long-term put options on Bitcoin at the average purchase price, yielding 3-5% per trade after paying premiums. Additionally, a custom API facilitates automatic e-transfers through bank logins for seamless buying and selling.

image showing gui for the game described

Nov 5, 2021

Java

JUnit

Rogue-like Game Built Using Java

The "Rogue Java Game" project is a Java-based game inspired by Rogue, developed using object-oriented principles. It supports custom room and item designs through a JSON file and includes features like saving and loading game states and changing player names. Development and testing were conducted on Java 11 and Windows 10. JUnit was utilized to implement rigorous testing of the program. Unit tests were written to ensure the functionality of various components, such as player movement, inventory management, and item interactions. This approach helped maintain code quality and reliability throughout the development process.

iamge showing homescreen of app described above

Oct 28, 2021

Node.js

Javascript

GPX File Manager Web-App with Custom C API

The GPX File Manager is a web application designed for managing .gpx files, commonly used in GPS devices. It features a custom memory-safe C API for parsing .gpx files and converting pertinent data into JSON format. The backend is built with Express.js, utilizing Node.js Foreign Function Interface (ffi-napi) for C function calls, and stores data in an SQLite database. The frontend is implemented with HTML, CSS, and Vanilla JavaScript for a user-friendly interface.