I am an AI and robotics researcher interested in graph-based modeling, networked autonomous systems, and multi-agent decision-making.
My current research focuses on using graph topology and network structure to model multi-robot environments, identify structurally critical components, and develop localized recovery and coordination strategies for multi-AGV systems.
Alongside research, I build end-to-end intelligent systems spanning robotic control, backend services, industrial automation integration, embedded systems, and full-stack web applications.
- 🎓 M.S. in Artificial Intelligence, Yeshiva University
- 🕸️ Working on graph- and topology-aware modeling for multi-robot systems
- 🤖 Researching multi-AGV coordination, task scheduling, and failure recovery
- 🏭 Experienced in real-world AGV deployment and industrial system integration
- 🦾 Building robotic control, visualization, and human–robot interaction systems
- 💻 Developing backend services and full-stack intelligent applications
- 📍 Based in the New York metropolitan area
- Graph-Based Modeling and Network Analysis
- Multi-Robot and Multi-AGV Systems
- Task Allocation, Scheduling, and Failure Recovery
- Multi-Agent Decision-Making
- Autonomous Systems and Motion Planning
- Robust and Adaptive Intelligent Systems
- Graph Representation Learning
- Reinforcement Learning
- Robot Learning
- Human–Robot Interaction
- Scientific Machine Learning
- Computer Vision
A graph- and topology-aware recovery framework for loaded task-execution failures in multi-AGV systems.
The warehouse environment is represented as a graph, where node criticality, alternative-path availability, traffic exposure, and operational dependencies are used to estimate the local impact of a failure.
The framework selectively identifies affected vehicles, tasks, and transfer resources for recovery while preserving unaffected vehicle schedules through reservation constraints.
The project includes warehouse topology modeling, failure injection, localized task reassignment, path recovery, comparative baselines, cross-layout experiments, and statistical evaluation.
Research Topics:
Graph Algorithms · Network Topology · Multi-Robot Systems · Multi-Agent
Coordination · Task Scheduling · Failure Recovery · Combinatorial Optimization
Designed, deployed, and integrated multi-AGV automation systems in industrial warehouse environments.
The systems connected warehouse management and control platforms with robot control systems, PLC-controlled equipment, databases, industrial networks, and heterogeneous AGV fleets.
Key responsibilities included:
- Designing API-based communication between WMS/WCS, RCS, PLCs, and AGVs
- Supporting task dispatching and real-time fleet coordination
- Investigating scheduling and execution behavior under warehouse constraints
- Identifying bottlenecks in task allocation, communication latency, and stability
- Debugging networking, databases, ports, logs, and hardware–software interfaces
- Translating industrial requirements into scalable system-level solutions
The experience included more than ten industrial projects and AGV fleets with up to 18 vehicles.
Engineering Topics:
Multi-AGV Coordination · Industrial Automation · WMS/WCS/RCS Integration ·
Task Dispatching · System Reliability · API Development · Cross-Layer Debugging
Developed an end-to-end robotic control platform integrating embedded actuation, backend services, browser-based remote control, and real-time three-dimensional visualization.
The system includes:
- Motor and servo control using PWM and GPIO
- OLED display and physical-button local control
- Automatic USB serial-device detection
- JSON-based robotic command transmission
- Modular Flask API and request-routing architecture
- Real-time browser-based control interfaces
- Dynamic status and gauge visualization
- Three.js-based 3D robotic-arm visualization
- Automated startup using Bash scripts and crontab
The platform supports both standalone and remote robotic operation and provides an extensible foundation for robotics control and human–robot interaction research.
Engineering Topics:
Robotics Software · Embedded Systems · Flask · REST APIs · Serial Communication ·
Three.js · Real-Time Systems · Human–Robot Interaction
Developed reinforcement learning agents for the Kaggle ConnectX environment, achieving a Top 9 worldwide ranking.
The project explored sequential decision-making in a competitive multi-agent environment, including:
- Value-based decision strategies
- Policy optimization
- Sparse-reward learning
- State-representation design
- Action-selection mechanisms
- Strategy evaluation against competitive agents
Research Topics:
Reinforcement Learning · Sequential Decision-Making · Multi-Agent Systems ·
Policy Optimization · Competitive Environments
Proposed a PDE-informed feature-engineering framework that integrates dynamical-system information into Gradient Boosting models.
Physically meaningful features, including spatial gradients and temporal derivatives, were extracted from dynamical simulations to improve predictive robustness and interpretability under sparse and noisy data.
Venue: IEEE Conference on Artificial Intelligence, 2025
Proposed mutual-information-based techniques for reducing redundancy among features in machine-learning models.
The framework decreases inter-feature dependency and incorporates Weight of Evidence transformation to improve feature representation, interpretability, and predictive performance.
Venue: IEEE International Conference on Consumer Electronics, 2025
Developed an adaptive deep-learning framework combining class-aware feature engineering with dynamical-system modeling.
The framework was designed to improve robustness under changing data distributions and support model adaptation without complete retraining.
Venue: IEEE SoutheastCon, 2025
Developed a modern web platform for an HVAC business, with an emphasis on responsive design, clear service presentation, customer-facing workflows, and usable interfaces across desktop and mobile devices.
The project includes:
- Responsive landing pages
- HVAC service presentation
- Customer contact and request workflows
- Scheduling-related interfaces
- Quotation and service-information components
- Reusable frontend components
- Mobile-friendly navigation and page layouts
Engineering Topics:
Full-Stack Development · Responsive Web Design · JavaScript · HTML · CSS ·
Frontend Architecture · UI/UX
My industrial experience connects academic research with real-world autonomous systems. I have worked across multiple layers of AGV systems, including robot control, task scheduling, system integration, databases, communication networks, and industrial interfaces.
This experience motivates my research in robust multi-robot coordination, failure recovery, graph-based environment modeling, and practical autonomous systems.
Served as a teaching assistant for Natural Language Processing, supporting students with:
- Classical machine-learning methods for NLP
- Text representation and classification
- Sequence labeling and text generation
- Transformer-based architectures
- BERT fine-tuning
- LLaMA prompting
- Retrieval-Augmented Generation
- LLM agent workflows
Graph Modeling · Network Analysis · Multi-AGV Systems · Multi-Robot Coordination · Motion Planning · Task Scheduling · Failure Recovery · Task Allocation · Industrial Automation
PyTorch · TensorFlow · Keras · Scikit-learn · Reinforcement Learning · Feature Engineering · Computer Vision · Scientific Machine Learning · Data Analysis
Flask · REST APIs · Serial Communication · Real-Time Control · Three.js · WMS/WCS/RCS Integration · PLC Integration · SQL Databases · Industrial Networks
Python · JavaScript · HTML · CSS · SQL · C · C++ · Java · MATLAB
STM32 · Arduino · Raspberry Pi · PWM/GPIO · Motor Control · Servo Control · Sensor Integration · OLED Interfaces · Circuit Design
Git · Linux · Tableau · Power BI · SolidWorks · AWS · GCP · Azure
- 🏆 Top 9 worldwide, Kaggle ConnectX Reinforcement Learning Competition
I am currently interested in Ph.D. opportunities related to:
- Graph-based robotics and networked autonomous systems
- Multi-robot coordination and planning
- Robot learning and multi-agent decision-making
- Robustness and failure recovery in autonomous systems
- Graph representation learning for physical and robotic systems
I am also open to research and engineering opportunities in robotics, autonomous systems, machine learning, industrial AI, and intelligent software development.
- 🎓 Google Scholar
- 💻 GitHub