My name is Max, a 3rd Year PhD student at the University of Sheffield in the Pyne Lab where I help uncover the role of DNA shape in self- and protein interactions through automated image analysis pipelines. I studied Physics at the University of Leeds during my undergraduate and Masters degree where I was introduced to Python and developed a love for scripting, coding and now software development.
-
π« How to reach me:
- mcgamill1(at)sheffield.ac.uk
- maxgamill(at)live.com
-
π Pronouns: he/him
-
β‘ Fun fact: In contrast to my coding work, I love adrenaline sports such as mountain biking, climbing, and skiing.
- π I'm currently working on: image analysis workflows for atomic force microscopy images of DNA using TopoStats:
- π± I'm currently learning about: image-based deep-learning models and how auto-encoders could help with shape analysis via transfer learning (private repo until published):
-
Created from scratch an interactive portfolio website showcasing early work such as; a simple feed-forward neural network to classify hand-drawn numbers trained on the MNIST dataset, a susceptible, recovered and infected (SIR) disease model, and monte-carlo simulations to model the magnetic moments of a 1D spin Ising model.
- A key challenge here was to create a frontend which could take client-side variables (and even drawing canvases!) and send these to the appropriate python scripts to be processed on the Raspberry Pi server.
- The project used: a html, css and javascript frontend, a Django and AJAX backend, hosted on a Ubuntu Raspberry Pi, equipped with an Apache gateway and port forwarding. (The webserver is currently down to free up the Raspberry Pi for other projects). Source code available on My Personal GitHub
-
(In progress) Creating from scratch an marketplace for personalised artwork / photo editing using a variety of algorithms and machine learning models (no generative models!). Once complete I hope that people would be able to apply funky filters to their own images and personalise them for cards / invitations such as those below: