About

I am Sabbir Alom Shuvo, a computational scientist and researcher pursuing a Ph.D. in Computational Analysis and Modeling at Louisiana Tech University. My research lies at the intersection of artificial intelligence, scientific machine learning, computational modeling, and advanced manufacturing, with a broader interest in developing intelligent computational frameworks for complex engineering and scientific systems. I am particularly interested in combining physical understanding with data-driven learning to improve prediction, interpretation, optimization, and engineering decision-making.

Currently, I work as a Graduate Research Assistant in the Additive Manufacturing, Mechanics, and Materials (AM³) Laboratory at the Institute for Micromanufacturing (IfM). My research focuses on advanced manufacturing processes including Laser Powder Bed Fusion (LPBF), Wire Arc Additive Manufacturing (WAAM), and Additive Friction Stir Deposition (AFSD). I investigate process–parameter relationships, material behavior, manufacturing variability, and defects while developing computational and machine-learning approaches for process prediction and optimization.

A major direction of my research is the development of physics-inspired and generative digital twins for manufacturing systems. My work includes forward process modeling, inverse parameter estimation, multimodal learning, probabilistic modeling, uncertainty assessment, and data-driven process optimization. In my LPBF research, I developed a bidirectional digital-twin framework capable of predicting melt-pool geometry, reconstructing cross-sectional melt-pool images, and identifying feasible manufacturing parameters for a desired geometry. This work integrates numerical, experimental, and image-based information through generative learning and was published in the Journal of Intelligent Manufacturing.

My broader research interests include scientific machine learning, generative and probabilistic modeling, inverse problems, small-data learning, uncertainty quantification, sensitivity analysis, computer vision, and multimodal data integration. I work with methods ranging from conditional variational autoencoders and convolutional neural networks to support-vector models, ensemble learning, reinforcement learning, Gaussian mixture models, and kernel density estimation. I also use statistical techniques such as multivariate analysis, correlation analysis, experimental design, feature importance, and sensitivity analysis to understand the mechanisms underlying complex datasets rather than treating machine-learning models purely as black boxes.

My research extends beyond additive manufacturing. I have developed computer-vision and probabilistic-analysis methods for wildfire smoke detection and density estimation, investigated machine-learning approaches for ammonia decomposition catalyst optimization, worked on intelligent recovery strategies for software-defined networks, and contributed to research in semiconductor technology, renewable energy, and applied data science. This interdisciplinary experience has shaped my broader goal of developing computational methods that can transfer across engineering domains rather than remaining limited to a single application.

Before joining the AM³ Laboratory, I worked as a Graduate Research Assistant in the Statistics and Machine Learning Laboratory at Louisiana Tech University, where I applied machine learning, computer vision, statistical analysis, sequential modeling, and reinforcement learning to interdisciplinary research problems. My work included data preprocessing, feature engineering, model development, hyperparameter optimization, cross-validation, and rigorous evaluation of model reliability. I have also served as a Graduate Teaching Assistant in General Chemistry laboratories, supporting undergraduate laboratory instruction, experimental activities, student supervision, and assessment. niw petition

I earned my M.S. in Computer Science from Louisiana Tech University and my B.Sc. in Computer Science and Engineering from Daffodil International University, Bangladesh. My background in computer science gave me an early foundation in programming, algorithms, software development, automation, robotics, and data systems, which I have progressively expanded toward scientific computing, engineering analysis, and computational research. Before beginning graduate research, I also trained as a Salesforce Development Trainee at Inovi Solutions, gaining practical experience in enterprise application development, structured data workflows, automation, testing, and debugging.

My research has resulted in peer-reviewed publications spanning advanced manufacturing, machine learning, sustainable energy, semiconductor technology, and applied computational science. These include research on a bidirectional physics-inspired digital twin for LPBF, machine-learning optimization of ammonia decomposition catalysts, thin-film solar-cell technologies, FinFET technology, and AI-era web security. I have also contributed to research on small-data machine learning for wire-arc additive manufacturing and computational studies relevant to emerging manufacturing systems.

Beyond research, I contribute to the academic community through peer review, interdisciplinary collaboration, student leadership, and mentoring. I have served in leadership roles including General Secretary of the Graduate Student Council and Treasurer of the Students’ Association of Bangladesh at Louisiana Tech University. I have also received academic recognition including the Entergy Corporation LP&L Endowed College of Engineering and Science Scholarship and Louisiana Tech doctoral tuition support.

My long-term research goal is to develop reliable, reusable, and generalizable computational intelligence frameworks that connect experimental data, physical knowledge, simulation, and artificial intelligence. I am particularly interested in advancing digital twins, inverse design, scientific machine learning, and multiscale computational approaches that can reduce experimental trial-and-error, improve process reliability, accelerate engineering design, and translate complex data into practical decisions.

Ultimately, I view artificial intelligence not simply as a prediction tool, but as a means of understanding, designing, and controlling complex physical systems. Whether I am studying manufacturing processes, analyzing scientific images, modeling material behavior, or developing intelligent engineering systems, my objective is the same: to transform data and computation into deeper scientific understanding and practical engineering impact.