I'm just a dude who works in energy markets, mostly doing optimisation modelling and data related things.
Did a bunch of maths and computer science at university, then actually learnt most of what I know on the job, and now applying it all to build some cool stuff in the energy space!
Here's a selection of some data and AI projects I've completed, either as part of university or outside.
The repository for my master's thesis, looking at using deep reinforcement learning strategic bidding of energy storage.
The goal was to detect the usage of sarcasm in online comments using pre-trained language models, as well as utilising and analysing their embeddings with traditional classification methods.
Developed game playing agents to play a multiplayer version of Pacman called Pacman Capture the Flag by UC Berkeley. It demonstrates the usage of PDDL, heuristic search and approximate Q-learning techniques.
Implemented collision free multi agent path finding algorithms in order for agents to operate in the Flatland environment, a train network simulator designed to model and address the problem of train scheduling and rescheduling.
Developed an autonomous game playing agent to play a made up board game called Expendibots, with functionality to play against other agents. Demonstrates usage of the Negamax search algorithm, along with Zobrist hashing for implementing transposition tables.
I am using this site as a way to experiment with Node, React, JavaScript and web design in general. Obviously, the easiest way to get started is to fork a template and start tweaking, learning as you go. The original source template for this website is linked below, as well as the repository hosting my current version.
I try to travel around a bit. Here's some photos I've taken to add some colour to the website.