Ferdowsi University of Mashhad
April 2026 – Present
Conducting research and experimental development in deep learning, context-aware neural networks, and intelligent analytical systems, with a focus on understanding model behavior and designing more effective feature representation strategies.
My current work explores convolutional architecture enhancement, multi-scale feature learning, adaptive feature fusion, and efficient global context modeling. A major direction of this research is investigating how neural networks can capture both fine-grained visual information and broader contextual relationships without relying on computationally expensive attention architectures.
This research has contributed to the design and experimental development of Med-AGCNet, a context-aware convolutional architecture built around the Adaptive Global Context Block (AGCB). The work includes architecture design, comparative model evaluation, component-level ablation analysis, error inspection, and Grad-CAM-based interpretation of prediction behavior.
Alongside deep learning research, I investigate AI-driven analytical methodologies for engineering and infrastructure systems, combining predictive modeling, simulation-based analysis, optimization, and interactive visualization to support technically complex decision-making workflows.
Research Focus: Deep Learning Architecture Design • Context-Aware Neural Networks • Multi-Scale Feature Learning • Adaptive Feature Fusion • Explainable AI • Applied AI Research
Behpouyan — Mashhad, Iran
March 2025 – September 2025
Worked on industrial data engineering and analytical systems focused on transforming complex operational information into structured, interpretable, and decision-oriented analytical workflows.
Designed data processing and transformation pipelines for engineering analysis and contributed to predictive modeling workflows aimed at investigating operational behavior, performance patterns, and infrastructure-related conditions.
Conducted model evaluation, error analysis, and comparative analytical assessment to examine prediction reliability and identify limitations across changing operational scenarios. Developed interactive dashboards and visualization workflows for trend analysis, engineering performance monitoring, and interpretation of analytical outputs.
Applied structured preprocessing and feature analysis methodologies to improve the representation of operational patterns and support reliable machine learning-oriented analysis. Worked with SQL-based data environments and structured analytical systems to support engineering and infrastructure analysis workflows.
Translated complex analytical outputs into structured technical reports and interpretable engineering insights, supporting communication between data-driven analysis and domain-specific engineering teams.
Core Focus: Data Engineering • Industrial Analytics • Predictive Modeling • Engineering Analysis • Data Visualization • Technical Reporting • Decision-Support Systems