Inductive Clustering For Lower Dimension Input Vectors
This article will discuss extending the non-inductive clustering algorithms to classify new instances and classify data samples with lower dimensions than those used for creating clusters.
Welcome to my portoflio
I'm a Data Scientist and a Machine Learning Engineer with Front-End Expertise
I have a degree in Computer Engineering from Cairo University and a Software Engineering background, especially front-end development. I love everything that has to do with computer science, programming, and machine learning.
During my studies, I had three internships as a Data scientist and a Machine Learning Researcher, In which I honed my skills in developing innovative solutions for various domains, including automotive, transportation, and energy industries.
I also got selected twice for The Google Summer of Code program, where I contributed to the open-source community. I gained much experience by working on ongoing projects that deliver real-world value.
If you're curious to dive deeper into my work, you can follow me:
CAD Software Engineer
PULSAR Microelectronics, Cairo, Egypt
February 2024 – Current
- Contributed to the main program codebase, focusing on fixing bugs, refactoring, and improving the overall code architecture.
- Designed and Implemented a highly interactive visualization tool for visualizing the different types of electronic simulations (AC, DC, etc ...) with different functionalities including adding markers, drawing horizontal and vertical lines on plots with intersections calculations, and customize plot appearances.
Frontend Engineer
NXT Gen, Cairo, Egypt
November 2023 – Current
Tackle and Talk Angular
- Built complex UI components for handling booking sessions and providers’ availability with different time zones.
- Maintaining and refactoring major parts in the application such as Auth pages and dashboards.
SuperFit Nextjs
- Implemented the main landing page for website with focus on responsive design and performance with nearly perfect lighthouse scores.
- Implemented user authentication with Authjs and Integrated payment flow with backend.
Data Science Intern
Synapse Analytics, Cairo, Egypt
June 2023 – September 2023
- Developed and implemented a machine learning model for cost prediction in transportation and logistics, achieving an average mean percentage error of less than 20 percent.
- Showcased proficiency in the complete machine learning life-cycle, starting from data cleaning and feature selection, and progressing through model development and deployment using Docker and FastAPI.
- Utilized MLFlow for managing multiple experiment versions, allowing for easy comparisons between different iterations of the machine learning model.
Software Engineer
Google Summer of Code at GCP Scanner
May 2023 – September 2023
- Designed and built a user-friendly and intuitive user interface for the GCP Scanner Tool using React, which is a tool that can help determine what level of access certain credentials possess on GCP.
- Successfully implemented advanced search and filtering functionalities, allowing users to efficiently navigate and extract relevant information from the scanner's output, improving productivity and ease of use.
- Seamlessly integrated the visualization tool with GCP Scanner, ensuring a cohesive ecosystem for users to analyze and act upon their cloud security scan results.
Data Science Intern
Raisa Energy LLC, Cairo, Egypt
August 2022 – October 2022
- Started a research project aimed at creating a machine learning model to accurately predict one of the geological features for oil and gas wells (BVHH).
- Successfully implemented an inductive clustering method to categorize oil and gas wells based on their geological attributes and published an article that covers my work
- Developed algorithms for predicting the shape of regions around oil wells (Drilling Spacing Units).
Software Engineer
Google Summer of Code at Emory-BMI
May 2022 – September 2022
- Contributing to the Eaglescope project on a tool for creating user interactive data visualization dashboards.
- Refactoring and improving the performance of the charts to maintain 60 FPS rendering for big-size datasets.
- Adding new multidimensional interactive visualizations to the project and implementing a user interface to create the dashboard without configuration files.
Deep Learning Research Intern
Compumacy
August 2021 – November 2021
- Trained a model for Driver Action Recognition using different deep learning architectures such as VGG16, EfficientNet, and MoviNet using transfer learning techniques.
- Implemented different deep learning models such as following word predication with LSTMs and implemented CNN architectures such as Efficientnet from scratch using Tensorflow.
Bachelor of Computer Engineering
Cairo University Faculty of Engineering
2018 - 2023
Grade: Excellence
Graduation Thesis: Tracking and Motion Prediction for Autonomous Vehicles
Discover the proficiencies that allow me build amazing things
Python, C/C++, JavaScript, TypeScript, SQL.
OOP, Data Structures, Algorithms, Problem Solving, Databases, GIT.
Tensorflow, Pytorch, OpenCV, Sklearn, Numpy, Pandas, Matplotlib.
HTML5, CSS3, Sass, Node.js, React.js, Next.js, Flask, FastAPI.
Reaching and Building a complete perception system that is capable of detecting, tracking, and predicting themotion of other vehicles on the road To be used in Autonomous cars
Doing an extensive literature review for already existing motion prediction techniques.
Provide accurate predictions for the future trajectories of the surrounding vehicles on The NuScences datasetwhile maintaining 15-FPS speed.
Utilizing the transformer architecture to build encoders for captioning both the spatial interactions between agents and the temporal interactions between historical trajectories of agents
Python, Pytorch, Linux, Git
A customizable social media platform mimiking tumblr which enables users to: post, follow blogs, chat with them and more.
In Dashboard, you can follow recommended blogs on the sidebar, see a featured post, create posts from the top panel or simply scroll through the latest posts from your followed blogs. Just once you reach the last post, more posts are loaded for you to see
React, Sass, Jest, Docker, Git, HTML, CSS, JavaScript
An autonomous rover with main objective is to do rescue operations, discover dangerous areas.
Main Implemented Features:
- Using a robotic arm to pick up objects
- Controlled remotely using a mobile application
- Sense the environment using different sensors and send the data to the mobile application
- Object detection and classification using a camera
Python, C++, ROS, Tensorflow, Linux, Flutter, Git
My blog where I write about machine learning, web development, and other things I find interesting.
This article will discuss extending the non-inductive clustering algorithms to classify new instances and classify data samples with lower dimensions than those used for creating clusters.