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:

Python OpenCV Deep Learning CNNs

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:

Python OpenCV TensorFlow/Keras CNN

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:

Python OpenCV Deep Learning Database Integration

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:

Python OpenCV Image Processing Deep Learning

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:

Python TensorFlow/Keras CNN

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:

Python TensorFlow/Keras CNN

~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 Consultation

Machine 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:

Python Scikit-learn Regression Models

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:

Python XGBoost Scikit-learn

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:

Python NLP Scikit-learn TF-IDF

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.