PROJECT_IMAGE
Retail Buyer Segmentation
End-to-end ML platform combining K-Means clustering and six classifiers, achieving 99.55% accuracy with automated preprocessing and business analytics.
FlaskScikit-LearnKMeansLogistic RegressionXGBoost
Available for AI/ML opportunities
Abdelrahman Bakr Mabrouk Ghazy
Bakr Ghazy
AI Engineer · Machine Learning Engineer · Data Scientist · Software Engineer
Passionate about building AI systems, machine learning applications, LLM-powered solutions, and data-driven products that solve real-world business problems.
Ain Shams University·Computer Science & AI·Cairo, Egypt
PyTorch · TensorFlow
LLMs · RAG
Vision Transformers
About
Engineering intelligence, end to end.
I design, train, and ship machine learning systems — from classical models to modern LLM and RAG pipelines. Currently studying Computer Science & AI at Ain Shams University, I work at the intersection of research, data, and product.
I'm a Computer Science & AI student focused on turning research-grade models into production systems people actually use. My work spans computer vision, NLP, LLM applications, and full-stack ML platforms — with an emphasis on measurable business impact.
I care about clean data, honest evaluation, and thoughtful UX around model outputs. I move fluently between Python notebooks, PyTorch training loops, backend services, and container deployments.
8+
Shipped Projects
4
Internships & Roles
1st
Kaggle Placement
Focus areas
Machine LearningDeep LearningComputer VisionLarge Language ModelsRAGData ScienceBackend DevelopmentSoftware EngineeringDockerCloudGit
Skills
A stack built for shipping AI.
From low-level model engineering to production-grade backends, visualization, and team leadership.
Programming Languages
05PythonC++JavaSQLJavaScript
Frameworks
09FlaskStreamlitTensorFlowPyTorchScikit-learnPandasNumPyOpenCVHuggingFace
Tools
06GitGitHubDockerOracle DatabasePower BILinux
Machine Learning
13Deep LearningComputer VisionNLPLLMsRAGFeature EngineeringData AnalysisClusteringClassificationRegressionModel OptimizationGrid SearchCross Validation
Visualization
03MatplotlibSeabornPower BI
Soft Skills
06LeadershipProblem SolvingCommunicationTeamworkPresentationMentoring
Experience
Work, teaching, and volunteering.
A blend of engineering, leadership, and community impact.
Technical Instructor
IEEE Al-Azhar Student Branch
- Delivered technical sessions.
- Mentored students.
- Organized technical camps.
Volunteer
United Nations Habitat (World Urban Forum)
- International event operations.
- Team coordination.
- Visitor support.
Elite Internship Program
ASU Career Center
- Leadership
- Event Management
- Project Coordination
Machine Learning Internship
NTI
- Machine Learning
- Model Development
- Data Analysis
Selected work
Projects with measurable outcomes.
A mix of applied ML, deep learning research, LLM systems, and full-stack data products.
PROJECT_IMAGE
Emotion Detection System
Advanced facial emotion recognition models achieving 74.8% Kaggle accuracy and 1st Place using Vision Transformers and deep learning.
PyTorchVision TransformersViTDeiTConvNeXt
PROJECT_IMAGE
Vitamin Deficiency Prediction
Regression and classification pipelines for vitamin deficiency prediction using feature engineering and machine learning.
PythonXGBoostScikit-Learn
PROJECT_IMAGE
Lung Cancer Survival Prediction
Machine learning models predicting patient survival using clinical datasets.
PythonScikit-LearnPandas
PROJECT_IMAGE
Medical Diagnosis Expert System
Rule-based medical diagnosis chatbot with a Streamlit interface.
PythonStreamlitRule Engine
PROJECT_IMAGE
Billing System
Oracle database application for invoices, payments, customers, and transactions.
OracleSQLJava
PROJECT_IMAGE
Business Analytics Dashboard
Marketing analytics dashboard providing insights into customer behavior and revenue.
Power BISQLDAX
PROJECT_IMAGE
LLM & RAG Assistant
Retrieval-Augmented Generation applications using embeddings, vector databases, and LLMs.
LangChainEmbeddingsVector DBLLMs
Achievements
Numbers that back the work.
1st
Place — Kaggle Emotion Detection Competition
99.55%
Retail Buyer Segmentation Accuracy
74.8%
Vision Transformer Accuracy
91.42%
CNN Validation Accuracy
Education & Certifications
Learning, formally and constantly.
Ain Shams University
Faculty of Computer Science and Artificial Intelligence
DegreeBachelor's
ProgramComputer Science & AI
GraduationExpected
August 2025
Machine Learning for Data Analysis — Digital Egypt Youth (Score 95%)
NTI · Ministry of Communications and IT
View Certificate
Gallery
Moments from the journey.
Competitions, hackathons, IEEE, NASA Space Apps, UN Habitat, and conferences.
Contact
Let's build something intelligent.
I'm open to full-time roles, internships, research collaborations, and freelance ML work.