Gianluca Truda's Projects
An artificial neural net built in pure Python.
UCT 3rd year capstone project providing an audio capture, segmentation and labelling tool
Some bioinformatics tools I wrote using Python.
Simple bash script to handle sound preferences for bluetooth headset.
Some notes and materials I made at the UCT Crypto Workshop with Linum Labs
Using the Java Fork/Join Framework to implement parallelised Mergesort and Quicksort across architectures.
"This assignment has the objective of developing your understanding of the use of semaphores for controlling access to resources such as often occurs in an operating system."
"This assignment focuses on networked applications, where you will develop a client-server based chat application in groups of three students."
Webserver for generating a daily digest page from my Kindle highlights.
:wrench: .files, including ~/.macos — sensible hacker defaults for macOS
Team nano's code for the Entelect University Challenge 2018
Assignments for VU Masters course in Evolutionary Computing
A newsreader app that uses LLM embeddings to semantically prioritise arXiv and Github updates
Personal genetics code.
Composable tools for doing useful things with GPT-4 (from the command line).
An extremely simple AI that plays an almost-optimal strategy for hangman.
Hashcode 2017 for fsociety
Source code from team SUDO~NANO's Google Hashcode 2018 participation.
Script to generate abstract images for Instagram.
A collection of helpful tools for Java development as a University of Cape Town CompSci student.
Some quick setup files for basic Linux sessions (e.g. AWS instances / virtual machines / Raspberry Pis)
A memento mori
WIP: an ML ensemble that accommodates missing data
A minimalistic digital version of morning pages journal
A demonstration of security principles for peer-to-peer communications of sensitive content.
World's best Boerewors image classifier, built with ResNets and Transfer Learning.
Quant. trading with ML on Numerai
Parsely for Lexical and Syntactic Analysis
Real-time tool for exploring the relationships between PCA components and input features