Machine Learning Scientist & Engineer

I am a researcher and engineer specializing in Graph Neural Networks (GNNs), Large Language Models (LLMs), High-Performance Computing and Scalable AI Infrastructure.

I hold a Ph.D. in Computing and a B.S. in Computer Science. My expertise lies in bridging the gap between theoretical research and practical, scalable deployment—designing novel algorithms (Zero-Shot Learning, Temporal Graphs) and architecting the high-performance systems required to run them.

Download CV (PDF)

Research & Publications

My research focuses on structural node representation and zero-shot learning on graphs. I have authored multiple papers on detecting structural patterns and temporal dynamics:

  • LLM-GMP (2025): Large Language Model-Based Message Passing for Zero-Shot Learning on Graphs (Accepted at IEEE Big Data 2025).
  • 2FWL-SIRGN (2024): A Scalable Structural 2-dimensional Folklore Weisfeiler-Lehman Graph Representation Learning Approach (Accepted at IEEE Big Data 2024).
  • Temporal SIR-GN (2023): Efficient and Effective Structural Representation Learning for Temporal Graphs (Accepted at VLDB 2023).
  • Botnet Detection (2021): Detecting Botnet Nodes via Structural Node Representation Learning (Accepted at IEEE Big Data 2021).

Teaching & Leadership

Beyond research, I have served as a Graduate Teaching Assistant for upper-division and graduate-level courses, mentoring students in complex algorithmic concepts:

  • CS 534: Machine Learning 
  • CS 535: Large Scale Data Analysis 
  • CS 321: Data Structures 

Engineering & MLOps

Unlike pure theorists, I maintain the full-stack engineering skillset required to take models from notebook to production.

  • AI Security & Networking: Implementation of structural learning algorithms for network attack detection and real-time identification of infected devices within network traffic.
  • Computer Vision Systems: Development of automated image recognition pipelines for security and surveillance applications .
  • Full-Stack Deployment: End-to-end web application lifecycle management, including database administration (SQL/MariaDB), backend deployment (Docker/Node.js), and long-term site maintenance.
  • Scalable Infrastructure: Orchestration of high-availability Linux servers, multi-GPU compute nodes, and local inference clusters (Ollama/AnythingLLM).

Languages: Python, Java, C++, Rust, C, SQL.

Frameworks: PyTorch, TensorFlow, NetworkX, DGL.