Eduba

What technologies are having the biggest impact on IT right now? Large language models are the obvious answer, but the real impact is in the orchestration layer sitting beneath them. The shift from “how do I use this AI tool” to “how do I connect AI to my existing databases, rules engines, and human workflows” is where the actual transformation happens. Most organizations are still treating AI as a standalone product when it should be treated as a compiler layer that translates intent into operations across their existing infrastructure.

How do you ensure cybersecurity and data protection across systems? The honest answer is that most organizations are not doing this well with AI systems specifically. My background in Marine Corps cryptographic systems taught me that security is about controlling information flow at every junction, not just perimeter defense. With AI, that means understanding exactly what data touches the model, where outputs go, and who can access what. The 60/30/10 framework we use (60% traditional database, 30% rule-based logic, 10% AI) actually improves security posture because you’re minimizing the attack surface where unpredictable AI behavior can create vulnerabilities.

What trends are driving digital transformation? The commoditization of intelligence is the fundamental trend. What cost millions in custom development five years ago now costs $20 a month. But most organizations are misreading this, they think the trend is “adopt AI faster” when the real trend is “the ability to orchestrate multiple tools intelligently becomes the differentiator.” The companies thriving are not the ones with the most AI tools but the ones who understand which tool solves which problem.

How are you managing cloud infrastructure and scalability? Our technical architecture runs on Azure and AWS depending on client needs. David McDermott, our Chief Architect, built EdubaWare on a stack designed for flexibility rather than scale for its own sake, using managed services (PostgreSQL, Redis, blob storage) rather than over-engineering for problems we do not have yet. The principle is that infrastructure should match actual load, not imagined future load. We have scaled implementations from 50 users to enterprise deployments, but we start simple and add complexity only when the data demands it.

What are the key skills in demand for IT professionals today? The ability to think architecturally about when to use AI and when not to. Technical prompt engineering matters less than understanding system design and knowing how to validate outputs. My Ethics Engine research showed that LLMs have measurable psychological patterns with 90% test-retest reliability, which means we can actually assess these systems rather than treating them as black boxes. The professionals who can evaluate AI behavior, not just deploy AI tools, will be the ones organizations need most.

How do you stay current in such a rapidly evolving field? I publish research. The Ethics Engine paper (arXiv:2510.11742) came out of genuine questions I could not answer with existing methods, so I built the methodology to answer them. Staying current means being in the conversation, not just reading about it. I also teach constantly, and nothing forces you to understand something deeply like having to explain it to skeptical executives or academics who will challenge your assumptions.

What’s your approach to innovation and problem-solving in IT? Start with the problem, not the technology. Most AI projects fail because they begin with “we should use AI” rather than “what are we actually trying to solve.” My approach is to have clients build their first solution with basic tools so they understand the problem deeply before we engineer anything sophisticated. Innovation is not about novelty, it is about finding the simplest solution that actually works and gets adopted. The 95% adoption rate we achieve comes from this discipline of building understanding before building systems.