2nd Year Computer Science Student at Cardiff University
Python Application Assistant
Hi, I'm Leo Coote, a second-year Computer Science student at Cardiff University. I'm passionate about artificial intelligence and technology, and I enjoy leveraging these tools to solve real-world problems.
Currently pursuing a degree in Computer Science at Cardiff University, I have built a strong foundation in programming, algorithms, and data structures.
I also hold a Microsoft Azure AI Fundamentals Ceritificate.
I completed a summer internship with the Down2Earth project, developing a PyPi package for forecasting water hazards and continuing as a part-time contributor.
My primary interest lies in AI, especially in applications involving large language models (LLMs).
In my spare time, I enjoy learning Japanese, reading literature and keeping fit.
I aim to deepen my understanding of AI technologies and work on projects that positively impact education and language learning.
Collaborated with the Down2Earth Team to develop a Python PyPi package for forecasting water hazards in the HAD region of Africa and other areas. Initially part of a university internship, later continued post-internship.
Developing an AI-driven language learning app using Flutter, where users practice languages by interacting with a large language model (Gemini). Firebase is integrated for authentication and cloud storage.
An AI model using the U-Net architecture created to detect cloud coverage from satillite images. Trained using PyTorch and open-source data. Check out the GitHub repo for more information.
Created a website showcasing the Boids algorithm by Craig Reynolds, simulating flocking behavior in animals like birds and fish. The project was the basis of a game built for a university competition, earning 1st place.
Trained a Generative Adversarial Network (GAN) to generate artificial celebrity images. The model was trained on a dataset of celebrity photos, creating realistic but non-existent individuals as part of a personal project.
A neural network framework built with core python (no numpy tensorflow etc). Implementing MLP's, Backpropagation and more from scratch.