Data Scientist & ML Engineer | Spatial Data Science & GeoAI | Deep Learning for Remote Sensing | MLOps

I focus on structural data science and deep learning architectures, building robust end-to-end machine learning pipelines ($O(N)$ efficiency), data ingestion engines, and secure automation interfaces that connect mathematical models with complex enterprise systems.

• Core AI/ML Stack: Python, Scikit-learn, NumPy, Pandas, PyTorch / TensorFlow, SciPy.
• Spatial Data Science: GeoPandas, Shapely, ArcPy (ArcGIS Pro API), Rasterio, PySal.
• MLOps & Infrastructure: FastAPI, Pydantic v2, Docker, Pytest, Ruff, Mypy, Makefiles, Git.

A high-performance Full-Stack Spatial Decision Support System (Agri-DSS) computing dynamic multi-attribute suitability matrices for regional agricultural planning.

• Vectorized Analytical Engine: Implements highly optimized, non-blocking asynchronous FastAPI endpoints powered by vectorized NumPy array slicing and GeoPandas dot products, achieving true $O(N)$ execution speed.
• Strict Constraint Validation: Utilizes Pydantic v2 advanced validation contexts to programmatically enforce Analytical Hierarchy Process (AHP) weight vector integrity ($\sum w_i = 1.0$) under strict floating-point precision.
• Swiss-Style Telemetry: Features a clean, high-contrast monochrome React/Vite/Tailwind dashboard tracking mathematical telemetry metrics (Mean, Max, Prime Zone distribution percentages) in real-time.

An institutional-grade Model Context Protocol (MCP) framework exposing exactly 100 specialized geoprocessing tools directly to LLM hosts and intelligent agents.

• Process Isolation: Built with a strict decoupled multi-process architecture (Async Core / Isolated Worker Subprocess) to guarantee runtime protection against environment blockages.
• Security Layer: Features a strict PathGuard sandbox enforcing prefix validation over database structures before any algorithmic execution occurs.

• Computer Vision for Remote Sensing: Formulating automated pipelines for agricultural pattern identification and urban object extraction from high-resolution multi-temporal satellite imagery using Convolutional Neural Networks (CNN) and U-Net segmentation models.
• Urban Resilience Forecasting: Engineering predictive spatial suitability matrices and long-term geometric resilience frameworks for horizon target lines using robust statistical models.

• Data Hub: muend.github.io
• SaaS Prototype: tarimsalkoridor.online
• Focus: Open to collaborative tracks involving production-grade Data Science, Spatial Machine Learning pipelines, and automated GeoAI systems architecture.