Hafiz Samiullah Email & Phone Number

Canada

Canada

Edmonton, Alberta, Canada

- Spearheaded MLOps practices for seamless deployment and maintenance of ML models in agricultural applications.- Achieved a 56x speed increase and reduced data scanned by 879,000x through efficient data lake query optimizations, significantly enhancing data retrieval performance.- Integrated a Gradio app for data lake ingestion and querying, streamlining data management and accessibility.- Developed CI/CD pipelines for efficient model updates and scalability using AWS, Docker, and Kubernetes.- Enhanced data profiling and artificial data generation techniques for more robust model training.- Implemented domain adaptation with GANs to mimic real-world challenges, optimizing model performance in dynamic agricultural environments.

- Managed end-to-end MLOps workflows, ensuring reliable and scalable AI solutions.- Developed and fine-tuned models for image segmentation in agricultural applications.- Conducted data-centric AI research with a focus on explainability and model optimization.- Mapped Google Earth data and reverse-engineered processes to enhance model accuracy.- Implemented JWT authentication to secure API endpoints and manage user access.- Diagnosed and prescribed fine-tuning strategies to improve model performance.

Regina, Saskatchewan, Canada

- Making GTFS real-time feed - Comparing GTFS Static data with a real-time feed- Analysis of data using Machine Learning

Regina

• Research Assistant (Machine Learning & Computer Vision) ‑ working on different real-time solutions for industrial problems related to computer vision.• Used Semantic Segmentation for making pixel-level decision maps for variable rate herbicide application.• Supervised damage detection in Canola plants using deep Learning, an end‑to‑end framework and model deployment• Disease classification in Plants using Deep Neural Networks• Weed type detection and localization in a multi-crop environment using supervised learning• Artificial data augmentation techniques, data enhancement using Conditional Generative adversarial network• Hands-on experience with data pre‑processing, regression, clustering, classification, support vector machines, decision trees, and dockers.• Teaching Assistant ‑ responsible for helping professors in marking assignments, taking exams, and conducting seminars.• Incorporated weekly one‑to‑one sessions to engage with students, reinforce their participation and hands‑on learning.

Master Of Applied Science, 90

- Dissertation: "A Deep Learning based approach for Canola and Weed Segmentation in Precision Agriculture"(Supervisor: Dr. Abdul Bais).

Bachelor Of Science (Gold Medalist), 3.65/4.0 Gpa