Mohamed Gad

Mohamed Gad

Hellos

I’m a Systems & Biomedical Engineering undergraduate at Cairo University. I have a strong interest in Software Engineering and Artificial Intelligence, with focus on Physics-Informed Machine Learning and Deep Learning. I aim to apply these skills to multimodal challenges.

Interests
  • ML & DL
  • Neuroscience
  • Computer Vision
  • Physics-Informed ML
  • Medical Imaging
Education

  • BSc in Systems & Biomedical Engineering, 2027

    Cairo University

Experience

 
 
 
 
 
Siemens EDA
Software Engineering Intern
Siemens EDA
March 2026 – Present
Part of the Calibre team.
 
 
 
 
 
University of Illinois Urbana-Champaign
Software Engineering Research Intern
University of Illinois Urbana-Champaign
May 2025 – August 2025
Developing LLMs for software testing related topics, namely mutation testing.
 
 
 
 
 
Cairo University Racing Team - Formula Student
Computer Vision Team Member
Cairo University Racing Team - Formula Student
September 2024 – May 2025
Developed a visual-based perception system for the autonomous racing car, with focus on objects detection, pose estimation, and 2D-to-3D scene reconstruction.
 
 
 
 
 
Emonovo
Brand Ambassador
Emonovo
April 2021 – February 2022
Promoted the brand through high school events, virtual meetings, and video content to grow the customer base.

Recent News

  • [2026.03] Started my Software Engineering long-term internship at Siemens EDA.
  • [2026.02] Joined the Open Science Community at Cohere Labs.
  • [2026.02] Our Journal Article Modalities and algorithms for generalized motor seizure detection and prediction: a scoping review has been published in Expert Review of Medical Devices.
  • [2025.10] Our paper Physics-Informed Neural Networks for Anisotropic Cardiac Activation Mapping has been accepted at ICCTA 2025.
  • [2025.05] Started my role as a Software Engineering Research Intern at the University of Illinois Urbana-Champaign.
    • See all

    Publications

    Ahmed Wesal, Farida Hesham, Mohamed Gad, Mostafa Abdelrahim, Nariman Ahmed, Nour Bahgat, Nour Zaki, Aliaa Youssef (2026). Modalities and algorithms for generalized motor seizure detection and prediction: a scoping review. In Expert Review of Medical Devices.

    DOI

    Mohamed Gad, Sherif El-Gendy, Mazen Atlam, Ahmed Mohamed (2025). Physics-Informed Neural Networks for Anisotropic Cardiac Activation Mapping. In ICCTA'25.

    DOI

    Academic Service

    Peer Reviewer

    Projects

    *
    All Machine Learning Computer Vision Image Processing OOP
    Latent Assumption Coder
    Latent Assumption Coder (LAC) is an automated evaluation pipeline that detects unspoken beliefs and silent decisions in AI agent Chain-of-Thought (CoT) traces. It uses an LLM-as-a-judge to segment execution logs, classify missed human-in-the-loop requests against an extensible codebook, and generate reports for an interactive timeline dashboard.
    Domain: AI Alignment, Agentic AI, LLM Evaluation, Interpretability | Tech: Python, Anthropic API, scikit-learn, Pydantic
    Code
    Latent Assumption Coder
    Tactic2Vec: Deep Metric Learning for Similar Soccer Plays Retrieval
    Tactic2Vec is a Deep Metric Learning Framework. It solves the problem of play retrieval by learning a dense vector representation where geometrically and tactically similar scenes are close to each other in the embedding space.
    Domain: Sports Analytics, Metric Learning | Tech: Python, PyTorch, FAISS, scikit-learn, Plotly
    Code
    Tactic2Vec: Deep Metric Learning for Similar Soccer Plays Retrieval
    FLAMeS Inference
    This package reduces the complexity of using FLAMeS, a model for multiple sclerosis lesion segmentation from FLAIR images, into one intuitive command.
    Domain: Neuroimaging, Medical Imaging | Tech: Python, PyTorch, nnU-Net, NiBabel
    Code
    FLAMeS Inference
    ParcVis: Neuroimaging Visualization Tool
    Python-based GUI tool for 3D and 4D neuroimaging data (NIfTI maps) visualization.
    Domain: Neuroimaging, Data Visualization | Tech: Python, Tkinter, NiBabel, NumPy, Matplotlib
    Code
    ParcVis: Neuroimaging Visualization Tool
    Physics-Informed Neural Network (PINN) for Cryosurgery Simulations
    PINN to solve the nonlinear PDE governing temperature distribution during cryosurgery simulations.
    Domain: Scientific ML, Physics-Informed ML, Biomedical Simulation | Tech: Python, PyTorch, NumPy, Pandas, Matplotlib
    Code
    Physics-Informed Neural Network (PINN) for Cryosurgery Simulations
    A 3D Simulator of the Patellar Reflex and Demyelination Effects
    Built a simulator that studies the effect of demyelination on neuronal signal propagation using the Hodgkin-Huxley model and 3D visualization using Unity. The project focuses on visualizing the knee jerk reflex. The simulation captures the delayed signal transmission weaker reflex caused by demyelination.
    Domain: Computational Neuroscience, Simulation | Tech: Unity (C#) | Python, SciPy, NumPy
    Code Demo Video Download
    A 3D Simulator of the Patellar Reflex and Demyelination Effects
    Cones Detection for Autonomous Driving Systems Using Classical Image Processing Techniques with OpenCV
    Domain: Computer Vision, Autonomous Driving | Tech: Python, OpenCV, NumPy
    Code
    Cones Detection for Autonomous Driving Systems Using Classical Image Processing Techniques with OpenCV
    Stocks Exchange Manager
    A Java-based platform utilizing Object-Oriented Programming (OOP) principles, designed for traders to buy and sell stocks, manage portfolios, and analyze market data within a simulated stock exchange environment.
    Domain: Software Engineering | Tech: Java, JavaFX, Maven | Concepts: OOP
    Code
    Stocks Exchange Manager
    Web server
    Web server built using Go (Golang) that demonstrates handling HTTP requests, serving static files, and processing form submissions. Domain: Backend, Networking | Tech: Golang
    Code
    Web server
    Nearest Neighbour and Bilinear Image Interpolation in Python
    This Python project implements two common image interpolation techniques; Nearest Neighbour Interpolation and Bilinear Interpolation. These techniques are useful for resizing images while preserving their quality to some extent.
    Domain: Image Processing | Tech: Python, NumPy, Pillow
    Code
    Nearest Neighbour and Bilinear Image Interpolation in Python
    Bow and Arrow Game using Java and Processing
    A game built using Processing and Java that tests your precision, speed, and strategic thinking!
    Domain: Game Development | Tech: Java, Processing5 | Concepts: OOP
    Code
    Bow and Arrow Game using Java and Processing

    ✏️ Blog Posts

    Why does Andrej Karpathy Use SVMs for Embedding Lookups?
    In this blog post, we’ll explore the Exemplar-SVM technique promoted by Andrej Karpathy. We’ll discuss the general embedding lookup process, what is exemplar-SVM, explain why this technique is effective, compare it to K-Nearest Neighbors (KNN), and walk through a brief code implementation.
    Mohamed Gad
    Jun 12, 2026 10 min read
    Why does Andrej Karpathy Use SVMs for Embedding Lookups?
    Sliding Window Inference in Large-Scale 3D Imaging Models
    In this post, I walk you through the solution to a huge memory bottleneck in 3D Imaging Models: the Sliding Window Inference technique. We will look at how tools like Patchly and MONAI implement this.
    Mohamed Gad
    Feb 2, 2026 8 min read
    Sliding Window Inference in Large-Scale 3D Imaging Models
    Hardware-based Encryption
    In the first section of this blog post, I preview a very interesting paper that proposes an intriguing approach to hardware-based encryption that relies on biomemristors. In the second part, I showcase my own project, which I worked on with a teammate, where we built our own hardware-based encryption and decryption circuit based on the XOR Cipher method.
    Mohamed Gad
    May 14, 2025 5 min read
    Hardware-based Encryption
    Biological Neurons VS AI Neurons (The Perceptron)
    The main goal of this blog is to exonerate biological neurons and show that a single biological neuron is much more computationally powerful and sophisticated than you might think.
    Mohamed Gad
    Mar 28, 2025 13 min read
    Biological Neurons VS AI Neurons (The Perceptron)
    Incorporating Synaptic Current Dynamics with Hodgkin-Huxley Model to Simulate a Network of Neurons
    Simulating a single neuron is an easy task, thanks to the Hodgkin Huxley model, but simulating a network of neurons is less trivial and less documented over the internet. In this blog, We’re going to simulate a simple network of neurons. We’re going to create a model and a simulation of two neurons that interact at a synapse.
    Mohamed Gad
    Jan 26, 2025 11 min read
    Incorporating Synaptic Current Dynamics with Hodgkin-Huxley Model to Simulate a Network of Neurons
    Introduction to Physics-informed Neural Networks
    Neural networks are well-known for their ability to learn patterns from data, helping us make predictions and decisions in a variety of fields. But what if we could make these networks even smarter by teaching them to follow the rules that govern the world around us?
    Mohamed Gad
    Oct 19, 2024 6 min read
    Introduction to Physics-informed Neural Networks
    Unveiling the Brain's Rainbow: A Look at the Brainbow Technique
    Imagine a world where each cell in our bodies could be uniquely colored, allowing us to track their movements and interactions with unprecedented detail. While this might sound like science fiction, a revolutionary technique known as Brainbow has brought us closer to this reality.
    Mohamed Gad
    Oct 2, 2024 7 min read
    Unveiling the Brain's Rainbow: A Look at the Brainbow Technique

    Talks

    Physics-Informed Neural Networks for Anisotropic Cardiac Activation Mapping, ICCTA 2025
    Presented our paper “Physics-Informed Neural Networks for Anisotropic Cardiac Activation Mapping” at The 35th International Conference on Computer Theory and Applications (ICCTA 2025).
    Nov 27, 2025 9:00 AM
    Slides
    Physics-Informed Neural Networks for Anisotropic Cardiac Activation Mapping, ICCTA 2025
    Cardiac Activation Mapping Using PINNs Presentation at the 12th URIC, Abu Dhabi University
    Presented our research project titled “Cardiac Activation Mapping for Atrial Fibrillation Diagnosis and Treatment Using PINNs” at the 12th URIC hosted by Abu Dhabi University.
    May 29, 2025 1:30 PM Abu Dhabi University (Virtual)
    Slides Video
    Cardiac Activation Mapping Using PINNs Presentation at the 12th URIC, Abu Dhabi University

    Contact

    If you have any questions, or if you just want to say hi, please feel free to reach out.