Muhammad Yasir Saleem

Delivered funded healthcare AI and biomedical R&D projects for global startups, research labs, and universities, providing production-ready and research-grade AI solutions. Designed and deployed EEG, ECG, and PPG analytics pipelines for real-time monitoring and clinical diagnostics, achieving 90%+ accuracy on multi-subject datasets. Developed YOLOv8, Swin Transformer, and Vision Transformer (ViT) based systems for disease detection, segmentation, and anomaly recognition. Built GPT-4 fine-tuned LLMs, RAG pipelines, and AI-powered research automation tools, reducing data analysis and reporting time by 60%. Deployed scalable end-to-end AI pipelines using FastAPI, and GCP Vertex AI, optimizing inference time by 40% through model quantization and parallelization. Collaborated with funded research teams and healthcare innovators to build AI solutions for diagnostics, signal analytics, and predictive healthcare systems.

Delivered contract-based AI solutions focused on computer vision and biomedical signal processing systems for real-world business and research applications. Designed and deployed deep learning models for image classification, object detection, and visual pattern recognition. Developed biomedical signal processing pipelines for ECG, PPG, and EEG signals, including signal preprocessing, feature extraction, and ML-based analysis. Integrated trained models into production environments using clean APIs and automated deployment workflows. Ensured reliable project delivery with clear documentation, reproducible experiments, and structured version control. Provided technical guidance on model architecture selection, performance optimization, and deployment best practices. Collaborated closely with stakeholders to translate project requirements into scalable, production-ready AI systems.

Delivered production-ready AI solutions for healthcare and industrial automation, transforming research concepts into reliable, scalable systems deployed in real business workflows. Designed, trained, and deployed computer vision and deep learning models for facial verification, product classification, and automated defect detection. Built end-to-end AI pipelines using CNN and Transformer-based architectures for high-accuracy image and video analysis. Fine-tuned GPT-based and diffusion models for synthetic data generation and data augmentation, improving model robustness and generalization. Integrated AI systems into production pipelines, enabling automation, performance monitoring, and measurable ROI improvements. Optimized model performance through hyperparameter tuning, data preprocessing, and validation workflows. Collaborated with cross-functional teams to translate business requirements into deployable, production-grade AI systems. Contributed to research implementations, benchmarking experiments, and reproducible AI workflows for academic and industrial projects.

Designed and developed machine learning models for predictive analytics in competitive research environments. Performed end-to-end exploratory data analysis (EDA), data cleaning, and feature engineering to improve dataset quality. Built classification and regression models using Scikit-learn and XGBoost for high-accuracy predictions. Applied dimensionality reduction and clustering techniques (PCA, k-means) for pattern discovery and data representation. Focused on model interpretability, statistical validation, and data storytelling for actionable insights. Improved model performance by 15–20% through systematic feature engineering and hyperparameter tuning.

Worked on real-world computer vision projects including image classification, object detection, and image segmentation systems. Designed and implemented CNN and RNN (LSTM) architectures for visual recognition and sequence-based learning tasks. Applied advanced image preprocessing, augmentation, and segmentation techniques to improve dataset quality and model generalization. Leveraged transfer learning by fine-tuning pre-trained models (VGG, ResNet, Inception) to achieve higher accuracy on limited datasets. Designed and trained Generative Adversarial Networks (GANs) for synthetic image generation and data augmentation. Built and optimized scalable deep learning pipelines using TensorFlow and PyTorch with modular, production-ready design. Gained hands-on experience in deploying vision-based AI systems and managing end-to-end deep learning workflows. Improved model accuracy by approximately 15–20% through systematic model optimization and data augmentation strategies.

Applied machine learning algorithms and statistical models to analyze and interpret real-world datasets for actionable insights. Performed data preprocessing, feature engineering, and data visualization to support exploratory data analysis and robust model development. Implemented supervised and unsupervised learning techniques for classification, regression, and clustering problems. Assisted in building, training, and evaluating ML models to improve prediction accuracy and analytical performance. Collaborated with mentors and senior engineers on projects focused on optimizing business operations through data-driven decision-making. Strengthened hands-on experience in Python-based data science workflows, model evaluation pipelines, and experimentation frameworks. Improved model performance by approximately 10–15% through effective feature engineering and algorithm tuning.

Supported NLP and classification pipelines for applied ML use cases. Assisted in training and evaluating ML models with hyperparameter optimization. Contributed to data preprocessing, feature engineering, and validation workflows. Collaborated with R&D teams to prototype early-stage AI systems.

Completed a Bachelor’s degree in Software Engineering from COMSATS University Islamabad (Sahiwal Campus) with a strong foundation in Computer Science, Data Structures, Algorithms, and Programming. Built a solid base in: • Object-Oriented Programming • Data Structures & Algorithms • Database Systems (SQL) • Software Design Principles • Problem Solving & Computational Thinking This academic background laid the foundation for my specialization in: Machine Learning, Computer Vision, Biomedical AI, and Data Science.