I am a PhD-level Data Scientist specializing in the development of robust AI systems for high-stakes industrial applications. My expertise lies at the intersection of Predictive Logistics, Computer Vision, and Structural Health Monitoring (SHM). I currently lead strategic demand modeling initiatives at GA Telesis, architecting end-to-end machine learning pipelines to solve complex aerospace supply chain challenges and infrastructure intelligence problems.
I maintain a specialized project allocation to balance industrial operations with technical innovation:
Architecting probabilistic forecasting frameworks for high-value aerospace components. My work focuses on:
- Intermittent Demand Modeling: Leveraging Temporal Fusion Transformers (TFT) and ensemble learning to manage zero-heavy demand patterns.
- Operational Optimization: Mitigating Aircraft-on-Ground (AOG) risks and lead-time volatility through advanced quantile regression and matrix profile analysis.
- Inventory Analytics: Developing scalable pipelines to optimize global inventory cycles and financial forecasting.
Advancing "Expert-in-the-Loop" systems for automated inspection and document intelligence:
- Saliency-Driven Inspection: Utilizing human eye-tracking data and Knowledge Distillation (ViT to CNN) to automate structural damage diagnostics.
- Document Intelligence: Building sophisticated document extraction pipelines utilizing DocTR, PaddleOCR, and DonUT frameworks.
- Instance Segmentation: Implementing Mask R-CNN architectures for high-precision concrete crack detection and structural integrity assessment.
| Domain | Expert Competencies |
|---|---|
| Machine Learning | Intermittent Demand, Quantile Regression, Matrix Profiles, LightGBM, XGBoost |
| Deep Learning | Vision Transformers (ViT), Saliency Prediction, Mask R-CNN, PyTorch, TensorFlow |
| Data Engineering | AWS SageMaker, Snowflake, MLflow, SQL, ETL Pipeline Design |
| Engineering / IoT | Signal Processing, Modal Analysis, LoRa, Wireless Smart Sensors, ROS |
My research career focuses on the application of deep learning to Resilient Infrastructure, specifically capturing human cognitive expertise to improve the reliability of automated diagnostics.
- Doctorate in Engineering: Specialized in Structural Health Monitoring and Artificial Intelligence.
- Research Focus: Domain Adaptation for SHM, Saliency-based damage detection, and Multimodal Data Fusion.
- Professional Roadmap: Currently pursuing the AWS Certified Solutions Architect – Associate to integrate deep learning models with cloud-native infrastructure.
- Aviation: Developing scalable probabilistic models for zero-heavy intermittent demand patterns.
- Infrastructure: Curating specialized datasets for concrete crack segmentation and structural integrity assessment.
- Software: Building modular, backtested financial and technical "Labs" for algorithmic performance evaluation.
"Translating high-frequency sensor data into actionable industrial intelligence."