Research
7+
Peer-Reviewed
Publications
7+
Journals
Reviewed
30+
Papers
Reviewed
4
Top-Tier
Venues
Vanier
Scholar
NSERC Canada
Graduate Scholarship
Gold
Medal
Governor General
of Canada · 2025
Research Themes
One core instrument. Two sharpening edges. Five threads of applied work.
Thread 01
Machine Learning for Risk and Resilience in Complex, Interdependent Systems
The core instrument — supervised learning frameworks built to capture high-dimensional, non-linear interactions where a system's components depend on each other. Predicting risk before it cascades, and engineering the resilience that follows. Sharpened by two edges, network theory and optimization, that make this machine learning rarer than a single-output, independent-features approach.
Thread 02
Deep Learning for Hazard Forecasting
State-of-the-art deep learning architectures applied to geospatial time-series data for province-scale wildfire prediction, urban flood forecasting, and real-time risk assessment — end-to-end pipelines designed for operational deployment in high-stakes environments, turning early hazard detection into resilience.
Thread 03
Graph-Theoretic Risk Modeling
The first sharpening edge — complex network analysis applied to model systemic risk and cascading failure propagation in interconnected systems. Formalizes how local disruptions escalate into system-wide failures, and identifies where in the network to intervene to build resilience.
Thread 04
Optimization-Augmented Machine Learning
The second sharpening edge — metaheuristic optimization applied to accelerate and improve model development. Replaces exhaustive search with intelligent, convergence-focused strategies, and carries predictions through to optimized, resilient decisions under real-world constraints.
Thread 05
LLM Behavioral Monitoring in Complex Systems
An active research direction — detecting reliability degradation, output drift, and failure modes in LLM-powered pipelines operating over interconnected data systems. Extending the same resilience-under-pressure thinking, from physical and project systems to AI systems themselves.
Published Research
Peer-Reviewed Publications
Full publication list available on Google Scholar · Web of Science · ORCID
Real-Time Machine Learning Risk Prediction for Autonomous Drone Operations
A closed-loop ML architecture continuously processes real-time flight, shipment, and environmental data to estimate accident likelihood and initiate mitigation responses — with parallel ANN models predicting multi-stakeholder consequence impacts on the public, businesses, and customers, closing the loop from prediction to resilience.
Supervised ML for Multi-KPI Prediction Under Interacting Systemic Risks
ML models trained on historical project data to simultaneously predict five interdependent performance indicators under the combined effect of interacting and systemic risks — treating multi-output prediction as a first-class modeling problem rather than an afterthought.
Ensemble Machine Learning for Real-Time Dynamic Risk Prediction
Ensemble algorithms trained on injury precursors to generate proactive, real-time safety risk predictions across work zones — moving risk management from reactive lagging indicators to predictive leading indicators.
Graph-Theoretic Machine Learning Pipeline for Cascading Risk Quantification
A three-stage pipeline where complex network theory quantifies team interdependencies as graph features, ML models predict project performance, and network-based and optimization-based mitigation strategies reduce systemic exposure and build resilience into project delivery.
Complex Network Theory and Metaheuristic Optimization for Systemic Failure Modeling
The origin of the network-theory edge: contractor interdependence quantified via weighted degree centrality on dynamic and static project networks — correlated with KPI degradation and mitigated through metaheuristic schedule reorganization. The network-theoretic and optimization foundation that later fuses with machine learning in the work above.
GA-Tuned ML and Multi-objective Optimization for Performance Prediction
Genetic algorithm hyperparameter tuning outperforms grid search across all ML models — predictions then feed NSGA-II multi-objective optimization to generate optimized schedules and risk registers that improve KPIs and build resilience under real-world constraints.
Research Philosophy
How I think about research
Pillar 01
Rigor Without Compromise
Research that cannot be reproduced is not research — it is speculation. Every model, pipeline, and result is held to the standard of reproducibility, statistical validity, and honest evaluation against strong baselines. Rigor is not a phase of the process; it is the process.
Pillar 02
Methods That Travel
A method that only works on the dataset it was built for has limited value. The goal is frameworks that generalize — where the core approach transfers across domains, scales to new data, and survives contact with real-world conditions that were not anticipated at design time.
Pillar 03
From Prediction to Resilience
Predictive accuracy is a means, not an end. Research that stops at a performance metric misses the point. The full value of a model is realized only when its outputs inform decisions and engineer resilience — which requires interpretability, uncertainty quantification, and integration with the constraints of operational environments.
Pillar 04
Reliability as a Design Constraint
Trustworthy systems are not built by accident. Reliability, resilience to distributional shift, and graceful failure under unexpected inputs must be designed in from the start — not patched in after deployment reveals the gaps.
Methods & Tools
Full Technical Stack
01
Machine & Deep Learning
Core libraries for model development, training, and evaluation across classical and deep learning paradigms.
02
Graphs & Optimization
Graph-theoretic analysis and metaheuristic optimization for complex systems modeling and decision support.
03
MLOps & Production Pipelines
End-to-end automation of ML workflows from data ingestion through deployment, monitoring, and retraining.
04
Cloud & Infrastructure
Scalable cloud-native environments for training, serving, and orchestrating production machine learning systems.
05
LLM & Agentic Systems
Frameworks and retrieval architectures for building grounded, autonomous, and observable LLM-powered systems.
06
Geospatial & Computer Vision
Spatial data processing, satellite imagery analysis, and vision-based feature engineering for geo-aware ML.
07
Data, Statistical & Visualization
Languages, statistical methods, and visualization tools spanning the full data analysis pipeline.
Research Impact
Where the research is going
The next frontier of this research is systems that don’t just predict accurately — but predict reliably, explain themselves, and remain trustworthy as the world they model continues to change. From graph-based risk intelligence to LLM behavioral monitoring, the through-line is the same: building AI that holds up under pressure, at scale, in production.