Hello, I'm
MSc in AI and Electrical Engineer specializing in AI/ML, Deep Learning, and IoT. Experienced in CNNs, RNNs, Transformers, and Feature Engineering, with hands-on deployment of real-time AI-IoT solutions on Edge Devices.
I am a Machine Learning Engineer focused on building practical AI and Deep Learning systems and integrating them with IoT and embedded platforms, including Arduino, ESP8266/ESP32, Raspberry Pi, and Jetson Nano. My work covers the full lifecycle from data preparation and feature engineering to model training, optimization, and deployment in connected and edge-based environments.
I develop computer vision and time-series solutions using CNNs, RNNs, LSTMs, Transformers, and YOLO, with experience in PyTorch, TensorFlow, and OpenCV. I work on optimizing models for real-time inference through quantization, performance tuning, and deployment on resource-constrained devices.
In parallel, I design and deploy complete AI-IoT systems by integrating sensors, communication modules, and cloud connectivity. My project work includes smart agriculture, water quality monitoring, Automation Systems, Road Management System, and intelligent monitoring platforms, where I combine embedded systems, edge AI, and cloud-based architectures to deliver reliable, production-oriented solutions.
Years Experience
Projects Completed
Publications
Designed, trained, and deployed end-to-end AI/ML/DL systems integrating CNNs, RNNs, Transformers, and YOLO for real-time edge inference on Raspberry Pi and Jetson Nano. Developed a comprehensive chemical water quality monitoring device that integrates multiple sensors with Raspberry Pi to measure water parameters, sending real-time readings and GPS locations to cloud databases and dashboards. Built a bacteria detection AI device by collecting experimental datasets, training and optimizing multiple deep learning models, and deploying the final system on Raspberry Pi for accurate in-field detection. Delivered both devices to organizations across KPK and Punjab under the GSMA project, supporting daily monitoring of chemical and biological water quality. Applied advanced ML/DL techniques including data preprocessing, feature engineering, hyperparameter tuning, model evaluation, quantization, pruning, and model compression for edge deployment. Built scalable ML pipelines for automated training, validation, and continuous model improvement to ensure robustness and reliability in real-world conditions. Integrated AI systems with cloud APIs, databases, and real-time dashboards for analytics, visualization, and monitoring. Collaborated with cross-functional teams to develop production-ready AI-IoT systems following software engineering best practices.
Developed an Industry 4.0–driven AI agriculture system integrating IoT, Machine Learning, and Deep Learning for intelligent crop scouting across incubator, greenhouse, and open-field environments. Designed and automated a smart plant incubator with environmental and soil sensors, camera modules, and closed-loop control systems driven by data-driven growth models. Built large-scale multimodal datasets (sensor time-series + plant images) covering complete plant lifecycles for multiple crops. Applied Machine Learning models for yield prediction and growth trend analysis using environmental and soil data. Developed and trained Deep Learning and Computer Vision models (CNN-based) for plant health assessment, growth stage classification, and food quality detection. Performed feature engineering, data preprocessing, model evaluation, and optimization for reliable real-world deployment. Integrated trained AI models into real-time dashboards for continuous monitoring, analytics, and comparative performance evaluation. Conducted cross-environment comparative analysis using AI-based insights to quantify the impact of controlled and natural growth conditions.
Contributed to a Safe City AI initiative, developing computer vision–based road and traffic management systems, including vehicle detection, counting, license plate recognition (ALPR), facial recognition, and person counting. Collected, curated, and annotated large-scale image and video datasets, and trained Machine Learning and Deep Learning models (CNN- and YOLO-based) for accurate real-time inference. Performed data preprocessing, augmentation, model evaluation, and optimization to improve detection performance under diverse urban conditions. Designed and deployed IoT-based environmental monitoring devices, integrating multi-parameter sensors with cloud servers and real-time dashboards. Created mechanical designs using SolidWorks and produced functional enclosures through 3D printing for field deployment.
Developed an autonomous mobile vacuum robot for smart trash collection using Raspberry Pi and Arduino. Implemented LIDAR-based SLAM and ROS2 navigation for real-time mapping, localization, and obstacle avoidance. Developed computer vision–based trash detection using OpenCV, enabling the robot to identify waste and trigger real-time alert/notification mechanisms. Integrated multiple sensors and actuators for autonomous operation, control, and environment perception. Validated system performance through Gazebo simulations and RViz visualizations, optimizing navigation and detection accuracy.
Deliver end-to-end AI/IoT solutions integrating ML models, GUI, Deep Learning, embedded devices, and cloud services, managing projects independently.
Pursuing advanced studies in AI/ML, Deep Learning, and their applications in real-world systems.
Specialized in electrical communication systems, embedded systems, and signal processing.
End-to-end expertise in Artificial Intelligence, Machine Learning, and Deep Learning — from research and model development to production-grade deployment on edge, embedded, and robotic platforms.
Specialized knowledge and skills across AI/ML domains
Areas of focus and ongoing research
Advancing state-of-the-art ML/DL algorithms, model architectures, and optimization techniques for real-world applications.
Applying AI/ML to healthcare challenges, medical imaging, signal processing, and diagnostic systems.
Object detection, image segmentation, medical imaging, and advanced vision algorithms for autonomous systems.
Large language models, text understanding, generation, and multimodal AI systems.
Autonomous robots, intelligent control systems, and AI-driven automation solutions.
Edge AI, IoT architectures, smart cities, and distributed intelligent systems.
AI-powered security systems, anomaly detection, threat analysis, and secure AI systems.
Responsible AI, model interpretability, fairness, bias mitigation, and ethical AI practices.
Click any certification to view the verified certificate
DeepLearning.AI (Coursera)
Foundations of deep learning & neural networks.
DeepLearning.AI (Coursera)
CNNs for computer vision applications.
DeepLearning.AI (Coursera)
Hyperparameter tuning & optimization.
DeepLearning.AI (Coursera)
ML project design & best practices.
DeepLearning.AI & Stanford (Coursera)
Regression & classification models.
UC Irvine (Coursera)
IoT systems & embedded fundamentals.
UC Irvine (Coursera)
Embedded C & Arduino development.
University of Michigan (Coursera)
Data processing & visualization.
University of Michigan (Coursera)
Core Python data structures.
University of Michigan (Coursera)
SQL & database access in Python.
University of Michigan (Coursera)
APIs, web scraping & JSON.
University of Michigan (Coursera)
Python programming basics.
IBM (Coursera)
Overview of data science field.
University of London (Coursera)
HTML, CSS & JavaScript basics.
Professional AI/ML solutions tailored to your business needs
Design and deployment of custom machine learning and deep learning solutions. From data preprocessing to production deployment, I build scalable, high-performance AI systems for real-world applications.
End-to-end computer vision systems for detection, segmentation, and real-time video analytics. Specialized in YOLO, CNNs, and advanced deep learning-based vision pipelines.
Intelligent text processing, chatbots, and large language model integrations. I build scalable NLP systems to automate workflows and extract business insights.
Production-grade ML systems with automated pipelines, monitoring, and cloud deployment. From raw data to scalable, reliable AI in production.
A collection of AI/ML, Computer Vision, IoT, and Robotics projects showcasing real-world applications
Developed an autonomous mobile vacuum robot for smart trash collection using Raspberry Pi and Arduino. Implemented LIDAR-based SLAM and ROS2 navigation for real-time mapping, localization, and obstacle avoidance. Developed computer vision–based trash detection using OpenCV, enabling the robot to identify waste and trigger real-time alert/notification mechanisms. Integrated multiple sensors and actuators for autonomous operation, control, and environment perception. Validated system performance through Gazebo simulations and RViz visualizations, optimizing navigation and detection accuracy.
Built an AI system for vehicle detection, classification, license plate recognition (ALPR), and traffic analysis with real-time dashboards for Safe City initiatives. Collected, curated, and annotated large-scale image and video datasets. Trained Machine Learning and Deep Learning models (CNN- and YOLO-based) for accurate real-time inference. Performed data preprocessing, augmentation, model evaluation, and optimization to improve detection performance under diverse urban conditions. Integrated with cloud servers and real-time dashboards for continuous monitoring and analytics.
Created DL models for plant disease detection, growth stage analysis, and yield prediction, integrated with IoT sensors for precision farming and smart agriculture systems. Developed Industry 4.0–driven AI agriculture system integrating IoT, Machine Learning, and Deep Learning for intelligent crop scouting across incubator, greenhouse, and open-field environments. Built large-scale multimodal datasets (sensor time-series + plant images) covering complete plant lifecycles. Applied Machine Learning models for yield prediction and growth trend analysis. Developed and trained Deep Learning and Computer Vision models (CNN-based) for plant health assessment, growth stage classification, and food quality detection.
Developed AI-IoT system for E. coli detection and chemical sensor monitoring, sending real-time data via MQTT/HTTP to cloud dashboards. Developed a chemical water quality monitoring device: integrated multiple sensors with Raspberry Pi to measure water parameters, send real-time readings and GPS locations to cloud databases and dashboards. Built a bacteria detection AI device: collected experimental datasets, trained and optimized multiple deep learning models, and deployed the final system on Raspberry Pi for accurate in-field detection. Delivered both devices to organizations across KPK and Punjab under the GSMA project, supporting daily monitoring of chemical and biological water quality.
Automated attendance system leveraging real-time facial recognition for accurate and efficient employee or student tracking. Integrates with existing databases for seamless record keeping. Developed using deep learning models for face detection and recognition, ensuring high accuracy and real-time processing capabilities. System includes features for enrollment, verification, and automated attendance logging with timestamp and location tracking.
Trained and optimized 3D object detection models, achieving mAP=90%. Converted Pandaset to KITTI format, compared PointRCNN, PV-RCNN, SECOND, and CenterPoint for vehicle/pedestrian detection. Implemented comprehensive evaluation metrics and performance analysis for autonomous driving applications. The system enables accurate 3D bounding box detection and classification in point cloud data, essential for autonomous navigation and safety systems.
Designed CNN-based automated sorting system for quality inspection, reducing post-harvest waste. The system classifies fruits and vegetables based on freshness, quality, and ripeness using deep learning models. Integrated with conveyor systems for automated sorting and grading, improving efficiency in agricultural supply chains.
Trained YOLOv5 model for real-time detection across multiple weapon classes for surveillance applications. The system provides accurate weapon detection in video streams and images, enabling enhanced security monitoring. Implemented with optimized inference for real-time performance in surveillance systems.
Automated large-scale drying with sensors, IoT monitoring, and CV-based spoiled fruit detection. Integrated environmental sensors for temperature, humidity, and airflow control. Computer vision system automatically detects and removes spoiled fruits during the drying process, ensuring quality control and reducing waste in industrial fruit processing.
Implementation of Principal Component Analysis for dimensionality reduction and data visualization. Demonstrated effectiveness on complex datasets, highlighting key variances. Applied to feature extraction and data preprocessing in machine learning pipelines, reducing computational complexity while preserving essential information.
Real-time lane detection system for autonomous vehicles, identifying and tracking lane lines using advanced computer vision techniques. Essential for navigation and safety in ADAS (Advanced Driver Assistance Systems). Implemented using edge detection, Hough transforms, and polynomial fitting for robust lane line detection under various road conditions.
A computer vision project that stitches multiple overlapping images together to create a seamless panoramic view. Features robust keypoint detection and image warping algorithms. Implemented using feature matching (SIFT/SURF), homography estimation, and blending techniques to create high-quality panoramic images from multiple camera viewpoints.
Optical Character Recognition system that automatically creates line-level bounding boxes in scanned documents. Enhances usability by converting printed/handwritten text to machine-readable format with precise localization. Integrated with text recognition models for end-to-end document digitization and text extraction from images.
Multi-modal emotion detection from image and speech using ResNet50 and MobileNetV2. Detects emotions: Happy, Sad, Surprise, Fear, with real-time processing. Combines facial expression analysis with audio processing for enhanced accuracy. Applied in human-computer interaction, sentiment analysis, and behavioral research applications.
Developed an AI-based early flood forecasting system using hydrological and climate data, fine-tuned for real-time conditions in Kalam and Swat regions, providing advance flood warnings and early alerts. Integrated IoT sensors for water level monitoring, rainfall data, and weather patterns. Machine learning models analyze historical and real-time data to predict flood risks, enabling proactive disaster management and public safety.
Artificial Intelligence Applications and Innovations. AIAI 2025 IFIP WG 12.5 International Workshops. IFIP Advances in Information and Communication Technology, vol 754. Springer, Cham.
View PublicationPakistan Journal of Engineering and Applied Sciences (Submitted)
I'm always open to discussing new opportunities and interesting projects.