Applied AI systems built at the intersection of safety, governance, and real-world deployment — each tied to peer-reviewed research.
A safety-first agentic architecture that uses blockchain as an immutable audit ledger for perception–reasoning–action pipelines. Ensures verifiability and real-time override capability in autonomous AI deployments.
A dual-objective agentic AI system for simultaneous financial budget optimization and personalized nutrition planning. Leverages multi-step constrained reasoning with preference alignment.
Integration of agentic AI within digital twin frameworks for smart city management. Enables autonomous optimization of urban infrastructure with real-time simulation and governance overlays.
Risk-aware multi-agent deep reinforcement learning system for cloudburst prediction and coordinated disaster response, with explicit uncertainty quantification and emergency protocol generation.
Ethical governance principles and technical implementation for AI-driven cybersecurity in renewable energy IoT networks, with autonomous threat detection and human-in-the-loop override.
An agentic digital-twin framework for predicting and preventing customer churn in retail banking. Continuously simulates customer behavioral trajectories with explainable retention interventions.