Hossein S
- Research Program Mentor
PhD at University of Maryland
Expertise
Data Science, Machine Learning, Deep Learning, Computer Vision, Large Language Models, Natural Language Processing, Cloud computing, Chemical Engineering, Thermodynamic modeling, Process Modeling
Bio
Since 2018 towards the end of my Ph.D. and then during my two postdoctoral fellowships, I have moved towards modeling-related areas and in particular data science and AI-driven approaches such as deep learning for computer vision, natural language processing, MLOps, etc. Currently, I hold a Senior Data Scientist position in an international pharmaceutical company, developing various digital, data science, AI/ML solutions for different manufacturing processes. I have 5 peer reviewed journal publications in areas related to applying AI/ML methods to solve problems in chemical and pharmaceuticals industry. I am passionate about teaching and breaking the barrier against entering the data science and related areas specially for folks in STEM and life sciences. I have served as mentor for several undergraduate and high-school students during my Ph.D. and postdoc, and interns during my presence in the industry.Project ideas
AI engine for knowledge management in large organizations
A mature organization, such as large corporations, insurance companies, research labs, manufacturing companies, etc., typically gathers years of knowledge and information. This project is focused on creating a search engine that uses AI/ML methods such as natural language processing (NLP) and language models to manage and search through this vast amount of data. The student will : Learn about a common industry problem and its various aspects and considerations Learn about basic principles behind Natural Language Processing and Large Language Models and other AI tools Learn to use common AI/ML Python packages and libraries to implement the models Develop programming and coding skills The main outcome of this exercise would be a prototype in form of a working python-based application.
Deep learning for medical imaging diagnostic assistance
Medical imaging plays a crucial role in healthcare, aiding doctors in diagnosing and treating diseases. In this project, students will develop a deep learning model to analyze medical images and then build a simple application for users to interact with it. This hands-on experience will introduce students to both machine learning engineering and software development, giving them a well-rounded exposure to building AI applications in medicine. The student will: - Learn about how AI is used in medical imaging and its real-world applications. - Gain an understanding of deep learning principles and image analysis techniques. - Work with popular AI/ML Python libraries such as Keras. - Develop introductory software engineering skills by integrating the trained model into a functional application. - Build a working prototype that allows users to upload images and receive AI-based insights. By the end of the project, students will have a foundational understanding of both AI model development and application deployment, preparing them for future work in data science and AI areas.
A chemistry calculator app
Computational tools are essential for scientists and engineers to perform complex calculations efficiently. In this project, students will develop a software application to assist with common chemical calculations, such as determining the heat of reactions, equilibrium constants, etc.. This project emphasizes learning how to write code and develop applications that support scientific work, bridging the gap between programming and real-world problem-solving in chemistry. The student will: - Learn about fundamental chemical concepts and how they are applied in calculations. - Develop programming skills using Python or another suitable language. - Work with scientific libraries such as NumPy and SciPy for numerical computations. - Design and build a user-friendly interface for inputting data and displaying results. By the end of the project, students will gain experience in both coding and scientific computing, equipping them with valuable skills for future studies in STEM fields.
Exploring enzyme activity through computational modeling
Enzymes are essential biological molecules that drive chemical reactions in living organisms. In this project, students will explore the features of enzymes and their activity through computational modeling and data analysis. By leveraging coding and scientific computing, students will gain insights into enzyme kinetics, factors affecting enzyme activity, and how enzymes are used in biotechnology and medicine. The student will: - Learn about enzyme structure, function, and the principles of enzyme kinetics. - Develop programming and data science skills to analyze enzyme activity data. - Use Python libraries such as Matplotlib and SciPy to visualize and model enzyme behavior. - Explore how temperature, pH, and substrate concentration influence enzyme activity. By the end of the project, students will gain hands-on experience in using computational tools to study biological systems, combining data science with real-world applications in biochemistry and biotechnology.