Kasra Mojallal
AI/Software Engineer building production LLM, retrieval, and vision systems. Research in federated learning and LLM security.
This profile was put together from Kasra's public GitHub portfolio, and they haven't joined Vibetown yet, so reach them through their portfolio. Is this you? Contact us to update or remove it.
About
Kasra Mojallal works in AI engineering and ML & data science, based in Canada (MealLens Inc.). AI/Software Engineer building production LLM, retrieval, and vision systems. Research in federated learning and LLM security. Public projects on GitHub include FedMod (Vertical federated learning with multi-server additive secret sharing: privacy-preserving training without…), Breast Cancer Stage Prediction (This repository contains machine learning models for breast cancer tumor staging, leveraging SVM, Random…) and Llm Vulnerability Testing (Red teaming and vulnerability testing for large language models with PyRIT and locally hosted LLMs), mostly written in Jupyter and Python.
Projects 6
FedMod
Vertical federated learning with multi-server additive secret sharing: privacy-preserving training without heavy cryptography. MSc thesis, published at IDEAS 2025.
github.comBreast Cancer Stage Prediction
This repository contains machine learning models for breast cancer tumor staging, leveraging SVM, Random Forest, AdaBoost, XGBoost, and ensemble methods to provide accurate and reliable staging predictions. It features comprehensive model evaluations and demonstrates the potential of ensemble methods in improving diagnostic accuracies.
github.comLlm Vulnerability Testing
Red teaming and vulnerability testing for large language models with PyRIT and locally hosted LLMs.
github.comMachine Learning Concepts
In this repository, I practice different ML concepts and libraries. Examples: ada-boost, xg-boost, openai-gym, openai-stable-baselines, roc-auc plots and many more.
github.comDigit Recognition Mnist
This project is about digit recognition of the MNIST dataset with neural networks, but the NNs were implemented from scratch. All of the back-propagation and other mathematics were implemented.
github.comBootcamp Machine Learning
This is the content for "Introduction to ML Boot-Camp" that I teach at Hermes-Capital academy.
github.com