Enterprise architect and applied AI researcher with more than 23 years of experience in enterprise architecture, cognitive systems, intelligent process automation, and digital transformation.
My research focuses on Active Inference, Expected Free Energy, explainable AI, agentic AI governance, business process management, process mining, enterprise capability mapping, software modernization, and transaction integrity.
I develop practical frameworks that connect AI reasoning, business outcomes, governance, and production architecture to enable transparent, adaptive, and trustworthy enterprise decision systems.
Research domains
Research Interests
Active Inference and Expected Free Energy
Explainable and Agentic AI
Business Process Management and Process Mining
Enterprise Architecture and Capability Mapping
AI Governance and Responsible Adoption
Transaction Integrity and Software Modernization
Executable research demonstration
Enterprise AI Standards & Governance Portal
Explore an operational implementation of the Balanced Enterprise Architecture Framework for AI Adoption. The portal turns governance principles into searchable standards, risk-tier checkpoints, release evidence, security and ethics controls, and production deployment gates.
Standards, checkpoints, policy artifacts, and evidence templates for secure, ethical, and governable enterprise AI adoption.
Executable research demonstration
Active Inference Recommender Research Portal
Explore the reproducible implementation behind intrinsically explainable recommendation through Epistemic-Pragmatic Expected Free Energy decomposition, including the interactive POC, benchmark protocols, model baselines, ablations, evaluation metrics, and evidence mapping.
Reproducible experiments, explanation traces, ablations, smoke tests, and publication-ready result generation without fabricated benchmark output.
Executable research demonstration
Outcome-Centric API Policy Portal
Explore a working implementation of dimension-level API policy guidance and build-time validation gates. The portal exposes executed test logs, traffic-light decisions, dimension scores, failed gates, JSON evidence, Markdown reports, and JUnit results.
