Muhammad saad

Built a hybrid deep learning pipeline for solar panel defect detection using CNN-based transfer learning, handcrafted feature fusion, and ML classifiers, achieving 97%+ classification accuracy. Applied CLAHE and Median Filtering preprocessing to 6,060+ images and deployed a Flask-based diagnostic web application with real-time inference and confidence scoring.

Built a hybrid deep learning pipeline for solar panel defect detection using CNN-based transfer learning, handcrafted feature fusion, and ML classifiers, achieving 97%+ classification accuracy. Applied CLAHE and Median Filtering preprocessing to 6,060+ images and deployed a Flask-based diagnostic web application with real-time inference and confidence scoring.

Trained and benchmarked CNN, YOLO, SVM, Random Forest, Logistic Regression, and Naive Bayes models across computer vision and tabular classification tasks. Executed full ML workflow including data collection, feature engineering, model training, evaluation, and Streamlit deployment.

Trained and benchmarked CNN, YOLO, SVM, Random Forest, Logistic Regression, and Naive Bayes models across computer vision and tabular classification tasks. Executed full ML workflow including data collection, feature engineering, model training, evaluation, and Streamlit deployment.

Developed responsive websites and web applications using HTML, CSS, Bootstrap, JavaScript, and PHP.

Developed responsive websites and web applications using HTML, CSS, Bootstrap, JavaScript, and PHP.

Grade: 3.19/4.0

Pursuing Bachelor of Science in Software Engineering with a CGPA of 3.19/4.0.

Grade: 3.19/4.0

Pursuing Bachelor of Science in Software Engineering with a CGPA of 3.19/4.0.

Grade: 941/1100

Completed FSC with a score of 941/1100.

Grade: 946/1100

Completed Matriculation with a score of 946/1100.

Grade: 941/1100

Completed FSC with a score of 941/1100.

Grade: 946/1100

Completed Matriculation with a score of 946/1100.