Mohamed Rodani

Physics Student Studying Machine Learning & Building Projects

I'm Mohamed, a physics student based in London with a growing focus on machine learning and AI. I enjoy working at the overlap between maths, physics, and code — from implementing ML algorithms from scratch to building small end-to-end projects. Right now I'm focused on deepening my fundamentals, competing on Kaggle, and building a portfolio of projects that show real understanding, not just copied code.

View Projects Download CV

Physics|Machine Learning

Python Machine Learning NLP Scientific Computing

Currently focused on deepening my ML foundations and building real projects.

About

I'm a first-year Physics BSc student at Queen Mary University of London (QMUL) with a deep fascination for machine learning and artificial intelligence. My background in physics has given me a strong foundation in mathematics, problem-solving, and analytical thinking—skills that translate naturally to the world of data science and ML.

My career goal is to transition from aspiring quantitative analyst to machine learning engineer, focusing on deep technical understanding over superficial "vibecoding." I've been actively learning through online courses, building projects, and applying ML techniques to real-world problems. My hackathon achievements include securing 4th place at the WCIT Hackathon and 1st place in the Tesco Competition.

I'm actively involved in the QMUL Machine Learning Society - Quant Division, where I engage in discussions on quantitative trading strategies, machine learning applications in finance, and collaborative projects. I believe in learning by doing—every project is a step toward deeper understanding. Whether it's implementing algorithms from scratch, working with neural networks, or applying ML to physics problems, I'm committed to continuous growth and practical application.

Dedicated to mastering Machine Learning through consistent reps, projects, and deep understanding—not just copied code.

Projects

What I Am Working On

Physics-Informed Neural Network (PINN) Solver

A machine learning model that uses physics-informed neural networks to solve partial differential equations relevant to physics.

Physics-Informed Neural Network Architecture

Python PyTorch NumPy Scientific Computing

Kaggle Playground Series

Participating in monthly Kaggle Playground Series competitions to practice machine learning techniques, feature engineering, and model optimization on diverse datasets.

Python Machine Learning Kaggle Data Science

ML Mini Projects

Collection of machine learning mini-projects covering various ML techniques, algorithms, and applications for learning and experimentation.

Python Machine Learning Data Science

Completed Projects

Heat Pump Explained

Hackathon project explaining heat pump technology, combining physics principles with computational modelling and visualisation.

Python Physics Modelling

RYM Investment Framework

Quantitative investment strategy framework built for the UK Investment Challenge 2025. Includes portfolio data pipeline, performance metrics, and comprehensive documentation.

Python Quantitative Finance Data Pipeline

BERT Financial Sentiment Analysis

Fine-tuned BERT model (110M parameters) on 5,842 financial headlines achieving 100% test accuracy. Implemented complete training pipeline including data loading, model architecture, and evaluation metrics. Successfully reduced training loss by 44% through hyperparameter optimisation.

Python PyTorch Transformers NLP FinBERT

Experience

November 2025

Diverse AI Hackathon 2025

Built AI transparency engine for smart heat pump systems in a 5-person team, addressing trust barriers in home energy automation through pattern detection and natural language explanations
Developed rule-based classifier identifying 5 operational scenarios (tariff optimisation, VPP curtailment, predictive heating, efficiency anomalies) with 3-tier recommendation system
Integrated Claude Sonnet 4 API with <50 word constraint, converting technical metrics (COP, flow temps, grid signals) into actionable insights
Grounded solution in EoH dataset (742 UK installations, 2+ years monitoring data) for realistic COP/SPF simulation
Delivered Streamlit dashboard with live Q&A chatbot, data transparency features, and benefit quantification (£ + kg CO2)

2023

WCIT Hackathon - 4th Place

Team Member

Competed in hackathon team, developing innovative solution. Secured 4th place finish among competing teams through technical excellence and collaborative problem-solving.

2022

Tesco Group Competition - 1st Place

Team Member / Lead

Won first place in competitive challenge. Led key technical contributions and collaborated with team to deliver winning solution through innovative approach and strong execution.

2025 - Present

QMUL Machine Learning Society - Quant Division

Active Member

Active participant in Machine Learning Society's quantitative finance division. Engaging in discussions on quantitative trading strategies, machine learning applications in finance, and collaborative projects.

Education

2025 - Present

Physics Degree

Bachelor of Science in Physics

Skills gained from studying Physics include mathematical modelling, computational methods, problem-solving, and analytical thinking.

2022 - 2024

A Levels

Mathematics, Further Mathematics, Physics

Completed A Levels in Mathematics, Further Mathematics, and Physics, building a strong foundation in quantitative and analytical skills.