Project Portfolio
A curated collection of production-ready AI solutions across Computer Vision, Machine Learning, and Generative AI. Each project demonstrates real-world problem-solving, technical depth, and deployment expertise.
10+ Completed Projects
Portfolio includes real implementations across Computer Vision, Machine Learning, and Generative AI domains with proven delivery.
Production-Ready Focus
Each project emphasizes deployment, integration, and real-world performance—not just model training.
Full Tech Stacks
Every project includes detailed descriptions, technologies used, and key outcomes for transparent evaluation.
How to Navigate This Portfolio
Computer Vision & Deep Learning: Real-time detection systems, image classification, autonomous systems, and advanced visual processing projects.
Machine Learning & NLP: Predictive models, classification systems, and natural language processing applications with proven accuracy metrics.
Tech Stack Details: Each project lists specific technologies (Python, OpenCV, TensorFlow, etc.) so you can assess technical fit for your needs.
Have a Similar Project in Mind?
Contact Ahmed to discuss your computer vision, machine learning, or AI requirements.
Computer Vision & Deep Learning Projects
Real-world implementations showcasing production-ready computer vision solutions. Each project demonstrates technical depth, real-time processing capabilities, and practical business impact.
Self-Driving Car Simulation
Developed an autonomous driving system using Computer Vision and Deep Learning. Implemented lane detection, object detection, and tracking in real-time video streams. Simulated driving decisions based on visual input and improved accuracy through testing, preprocessing, and model optimization.
Tech Stack:
Real-time lane detection and tracking
Multi-object detection in video streams
Autonomous decision-making simulation
Face Mask Detection System
Built a real-time face mask detection system using CNN and OpenCV. Classified masked vs unmasked faces in live video streams with high accuracy. Applied preprocessing, augmentation, and real-time inference for production deployment in security and compliance monitoring applications.
Tech Stack:
Real-time face detection and classification
Data augmentation for robust performance
Live video inference and alert systems
Face Recognition Attendance System
Developed an automated attendance system using face recognition and deep learning. Extracted facial features and matched identities with high accuracy. Integrated database system for storing and managing attendance records. Reduced false recognition errors through model tuning and optimization.
Tech Stack:
Facial feature extraction and embedding
Identity matching with confidence scoring
Automated attendance database management
Smart Parking System
Designed an intelligent parking detection system using advanced image processing and deep learning. Detected available parking spaces using segmentation and object detection algorithms. Built real-time monitoring system for parking availability. Optimized detection pipeline for improved accuracy and reliability.
Tech Stack:
Real-time parking space detection
Segmentation and object detection
Availability monitoring and optimization
Cat vs Dog Image Classifier
Built a CNN-based image classification system for binary classification of cats and dogs. Implemented data augmentation and regularization techniques to improve model generalization. Achieved high accuracy on test datasets through careful model architecture design and hyperparameter optimization.
Tech Stack:
Binary image classification architecture
Data augmentation for robustness
Regularization to prevent overfitting
Handwritten Digit Recognition (MNIST)
Developed a CNN model for handwritten digit classification on the MNIST dataset. Achieved approximately 99% accuracy through careful architecture design and deep learning optimization techniques. Demonstrates proficiency in building production-grade classification models with state-of-the-art performance.
Tech Stack:
~99% accuracy on MNIST dataset
Optimized CNN architecture
Deep learning best practices applied
These projects demonstrate production-ready computer vision capabilities, from real-time detection systems to advanced classification models. Each project showcases technical depth, optimization expertise, and real-world applicability.
Ready to discuss a similar project or explore how these capabilities can solve your business challenges?
Request ConsultationMachine Learning & NLP Projects
Explore my work across predictive modeling, regression analysis, and natural language processing. These projects demonstrate breadth across different AI domains and real-world problem-solving beyond computer vision.
Student Performance Prediction
Regression model predicting student exam performance with feature engineering and multi-algorithm comparison.
Key Outcomes:
- • Regression model comparing multiple ML algorithms
- • Comprehensive feature engineering pipeline
- • Model evaluation and performance metrics
Tech Stack:
House Price Prediction
XGBoost and Gradient Boosting model with feature engineering, hyperparameter tuning, and RMSE optimization.
Key Outcomes:
- • XGBoost and Gradient Boosting implementation
- • Advanced feature engineering and selection
- • Optimized RMSE performance through tuning
Tech Stack:
Spam Email Classifier
NLP-based spam detection system with text preprocessing, tokenization, and TF-IDF features achieving 95% accuracy.
Key Outcomes:
- • 95% accuracy spam detection system
- • Text preprocessing and tokenization pipeline
- • TF-IDF feature extraction and classification
Tech Stack:
Full AI Delivery Stack
These machine learning and NLP projects demonstrate breadth across different AI domains. Combined with my computer vision expertise, I deliver production-ready solutions across the full AI spectrum — from data preprocessing and model development through deployment and optimization.
Have a project in mind?
Contact Ahmed to discuss your computer vision, machine learning, or generative AI needs.
Whether you need real-time object detection, predictive modeling, custom LLM applications, or production deployment — Ahmed can help assess which services best fit your requirements and deliver production-ready solutions.
Quick Turnaround
Free initial consultation to understand your project scope and timeline.
Custom Solutions
Tailored AI models built specifically for your business problem and data.
Production Ready
Deployed solutions with MLOps integration and ongoing support.