Who Am I?
I'm Ahmed Nezar, a Machine Learning Engineer at Siemens EDA & passionate student pursuing a degree in Computer Engineering Software Systems at Ain Shams University. Enthusiastic about AI, Data Science, ML, DL, and Automation with hands-on experience in the mentioned fields in addition to generative AI. Equipped with strong programming and analytical skills, and always eager to learn and grow. Aiming to develop groundbreaking AI/Software solutions that drive innovation and solve real-world problems.
Machine Learning
Deep Learning
Data Science
Software
My Skills
Python
Machine Learning (Scikit-learn)
Data Visualizations (Matplotlib/SeaBorn)
Deep Learning (PyTorch)
Data Analysis (NumPy/Pandas)
Web Development (Django)
Education
Expected 2026
Major in Computer Engineering & Software Systems
Thanwia Amma
Grade:84.39%
Ranked 3 on School (Maneret El Eman Language School)
Work Experience
Deep Learning Path Planning Engineer - Formula AI ASU Racing Team November 2023 - July 2024
- Worked on generating dataset in order to train a seq2seq model.
- Developed a transformer model where it works by taking sequence of cones & generating sequence of way-points that will be later used to construct a path for the car.
- Integrated with other modules as perception & SLAM in order test our module.
- The system details con be found on this Research gate article.
AI Engineering Trainee at Orange Digital Center March 2024- May 2024
- Implemented machine learning algorithms using c++ & utilizing the use of packages as mlpack & armadillo
- Implemented deep learning models as CNNs & neural networks using PyTorch in c++.
- Understood the mathematical intuition behind the algorithms.
Machine Learning Mentor at Cellula Technologies August 2023 - March 2024
- Responsible for machine learning mentoring & guiding new trainees in their task while constantly giving them feedback.
- Worked on machine learning models mainly supervised as Regression & Classification.
- Applied EDA & visualizations to the data while pre-processing the data-set for better working with the given data while mentoring the interns.
- Deployed Machine Learning models using Flask & Django in order to have a full Machine Learning Model.
AI Trainee at Telecom Egypt July 2023 - August 2023
- Understood Artificial Neural Networks & Applied with them using Python.
- Learned about Convolution Neural Networks & their architecture while knowing difference between CNN & ANN.
- Used MaskRCNN in a project where it was an object detection & segmentation.
- Utilized the use of CV2, TensorFlow & keras.
Recent Projects
Image Generation Using GANs
- Understood different GANs architecture as GANs, DCGANs & WGANs.
- Augmented some of the dataset by making some flips or rotations in the images to generate different variety of images.
- Worked on the generation of boot-sandal-shoes images using the conditional version of each GAN architecture.
- Made Comparative Analysis on the different architectures used & comparing them in terms of FID & the quality of images generated.
Super Mario Bros using Reinforcement Learning
- Pre-processed Mario environment & understood how rewarding systems work.
- Applied some scaling to the environment & stacked them into a vector.
- Used PPO Reinforcement Learning Algorithm to train the agent.
- The model successfully passes the first level of Mario & getting a total score of 21650.
Concrete Cracks Classifier
- Pre-processed concrete images in order to able to classify it.
- Utilized the use of different architectures as ResNet50, VGG16, & ResNet18.
- Customized fully connected layer of the pretrained models to be able to classify the data.
- Achieved a high accuracy of 99% while evaluating the three pretrained models.
SpaceX Data Science Capstone Project |
- Predicted if the Falcon 9 first stage will land successfully.
- Used Folium to visualise launch sites
- Created a web visualisation page using Plotly Dash
Object Detection & Image Segmentation
- Used MaskRCNN to segement & detect all objects in the video.
- Used MaskRCNN to segment & detect cars only in the video.
- Used CV2 & pixellib library to count the cars in each frame in the whole video.
News Classification Using Classical NLP Techniques
- Built a machine learning pipeline using CountVectorizer, TF-IDF, and RandomForestClassifier to classify news articles into categories like "World", "Sports", "Business", and "Sci/Tech".
- Developed a modular workflow covering data preprocessing, model training, evaluation, and persistence, with reusable functions for each step, ensuring scalability and maintainability.
- Generated detailed classification reports to assess model accuracy and saved trained models using joblib for future deployment, demonstrating a focus on practical usability and performance analysis.
Image Classification on Fashion Mnist
- Applied the concept of CNN on Fashion Mnist Dataset.
- Used activation functions as ReLU & softmax in the Fully connected layer in order to see the probabilites of each class in the output layer.
- Made the Model Have an accuracy of 91%.
Hand-Written number classification on MNIST Dataset
- Applied the neural Networks on MNIST Dataset.
- Made the model have an accuracy of 97%.
- Implemented activation fuctions & showed images of the data using matplotlib.
DHCP Server Implementation
- Developed a DHCP server and client simulation to dynamically allocate IP addresses, handle lease times, and manage IP pools. The server supports features like IP assignment, lease renewal, and handling DHCP messages.
- Built a modular system with separate components for the DHCP server, virtual client, and GUI for both server and client management.
- Created an interactive GUI using customtkinter for managing the DHCP server and client. The GUI allows users to start/stop the server, configure client modes and view real-time logs and available IPs.
VeriGuard-Verilog Lint
- Developed a Python-based tool to analyze Verilog code, detecting issues like non-full case statements, uninitialized registers, latch inference, arithmetic overflow, multidriven variables, and non-parallel case statements.
- The tool checks for unreachable code blocks, feedback loops, and missing sensitivity list signals in always blocks. It also validates case statements for completeness and parallelism, ensuring robust and efficient hardware design.
- Generates a detailed text-based report summarizing all detected issues, including module names, line numbers, etc...
ATM Verilog FSM
- Designed and implemented a functional ATM (Automated Teller Machine) system using Verilog, simulating key operations such as balance inquiry, cash withdrawal, and PIN verification. The project demonstrates the use of finite state machines (FSMs) to manage system states and transitions effectively..
- Thoroughly tested the system using testbenches to verify functionality under various scenarios, including incorrect PIN entries and insufficient balance cases.
- The project highlights strong Verilog coding practices and hardware simulation skills.
Ticket Reservation System
- Built a Ticket Reservation System using Electron, a framework for creating desktop applications with HTML, CSS, and JavaScript
- The system allows users to book, cancel, and view tickets for events, providing a user-friendly graphical interface for seamless interaction.
- Utilized local storage or file-based storage to persistently store user and ticket information. The system includes error handling for edge cases, such as duplicate bookings or invalid inputs, etc...
Educapedia IGCSE Web Application
- Applying Software Principles & MVT architecture.
- Used HTML/CSS & JavaScript for Front-End & used Django for the backend.
- Hosted the application on pythonanywhere.
- Used Incremental Approach when developing the application.
Home Appliance Control Project
- This project is an implementation of home control system using a tivaC board and connecting with a python desktop application.
RC car
- Used an Arduino to control the car by receiving Bluetooth signals.
- Developed a mobile application to control the car using the Bluetooth sensor.
- Controlled the speed of the car using the Arduino.
- Had a 7th place finish in Robolympics Competition held by Aviation.
Certificates
July 20, 2024 | FS-AI
Formula Student AI
Participated in Formula Student AI in UK for autonomous driving vehicles specialized in Deep Learning Planning
July 1, 2024 | AI Engineering Diploma
AI Engineering Diploma
Took an AI Diploma where I understood core concepts of AI & implemented them in C++
January 31, 2024 | AI Engineering
IBM AI Engineering
Professional certificate containing 6 courses that explained computer vision fundementals, ANN, CNN, LSTM, RNN, DBNs, PyTorch, TensorFlow & Keras
September 4, 2023 | Data Science
IBM Data Science
Professional certificate containing 10 courses that From EDA & Preprocessing to Machine Learning Models
July 1, 2023 | Machine Learning
Machine Learning
Professional Certificate of Machine Learning Offered by DeepLearning.ai where it explained supervised, unsupervised & reinforcment learning throughout the 3 courses
July, 2023 | Voulnteering
Certificate of Recognition for instructing
Certificate given from IHUB for preparing content and explaining as an instructor.
July, 2023 | Voulnteering
Certificate of Recognition for Organinzing
Certificate given from IHUB for Organinzing the IChallenge23.
May, 2023 | Voulnteering
Certificate of Achievemnt STP23
Appreciation certificate after being a successful moderator through the year.
November 4, 2022 | Python
Python Data Structures
Learned about data structures for python programming language, this course was offered by University of Michighan.
September 27, 2022 | Python
Programming for Everybody (Getting Started with Python)
Understood the basics of python, this course was offered by University of Michighan.
July 20, 2022 | C Programming
C for Everyone: Programming Fundamentals
Applied concepts related to arrays & pointers, functions & storage classes, logic operators & various question types, lexical elements & data types, this course was offered by University of Santa Cruz
Febrarury 2022 | Python & Arduino
Python & Arduino Workshop
Used Python with Arduino & learned about how arduino works.
