Sepehr Mohammady
Intern - Tiny Machine Learning Reasearch @University Of Genoa
Signup · Get unlimited contacts
WORK HISTORY
Intern - Tiny Machine Learning Reasearch @University Of Genoa
IT
Evaluated and compared multiple machine learning models (including CNN and DCNN) for ToF image classification, applying advanced techniques such as partial binarization with the Larq framework. • Achieved 92.8% test accuracy with a resource-efficient model converted to TFLite, demonstrating strong potential for deployment on STM32 microcontrollers. • Developed and optimized a TinyML speech command classifier using Binarized Neural Networks (BNNs) with MFCC+delta features. • Processed the Google Speech Commands v0.02 dataset (10 classes), improving accuracy from ~30%(raw audio) to ~93.4% using extracted features. • Conducted an ablation study comparing Full Precision, QAT, and BNN models • Adapted a state-of-the-art NAS-BNN framework to classify CIFAR-10 images, showcasing your expertise in neural architecture search, custom data preparation, and pipeline orchestration.
EDUCATION
Islamic Azad University
Master of Engineering (M.Eng.), Information Technology
Naghshe Jahan
Associate's degree, Electrical and Electronics Engineering
University of Genoa
Doctor of Philosophy - PhD, Planning and Decision Methods
University of Genoa
Master's degree, ENGINEERING TECHNOLOGY FOR STRATEGY (AND SECURITY)
Academic Center for Education, Culture and Research
Bachelor of Technology (B.Tech.), ICT
ABOUT SEPEHR MOHAMMADY
As an IT Specialist, Data Scientist, and Machine Learning Researcher, I am passionate about leveraging technology to solve complex problems and optimize decision-making through data-driven insights. With a background in engineering technology for strategy and security from the University of Genoa, I have gained expertise in data analysis, machine learning, and IT strategy, enabling me to develop scalable solutions across various domains. I am conducting Tiny Machine Learning research, focusing on optimizing neural networks for deployment on resource-constrained devices. My experience includes refining models for speech recognition, image classification, and neural architecture search, achieving high accuracy while maintaining computational efficiency. Additionally, my proficiency in Python, MySQL, TensorFlow, PyTorch, and advanced analytics tools allows me to drive innovation in data science and IT infrastructure. Beyond research, I have held leadership roles in information architecture, content management, and digital strategy, working across industries in the USA, Algeria, Iran, and Italy. From managing complex IT ecosystems to analyzing business intelligence metrics, I thrive in multidisciplinary environments that require both technical expertise and strategic thinking. My goal is to bridge the gap between AI-driven analytics and practical business applications, delivering data-powered solutions that enhance efficiency and drive growth. I am eager to collaborate with professionals in data science, AI, and IT strategy to create impactful and intelligent solutions.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.