Integrated Information User Support Environments

AI-enabled knowledge systems for complex engineering, decision support, and enterprise data management

Integrated Information User Support Environments is a newly developing consulting and research organization focused on helping teams use artificial intelligence, semantic knowledge graphs, and human-centered software tools to improve complex knowledge work.

Our work builds on long careers in artificial intelligence, systems engineering, knowledge-based systems, software engineering, and enterprise consulting. We are especially interested in systems where critical knowledge is distributed across documents, models, requirements, tests, project artifacts, communication channels, and the tacit expertise of experienced engineers, architects, developers, testers, integrators, and program managers.

Who We Are

Robert Neches has spent his career applying artificial intelligence, knowledge-based systems, and advanced software concepts to problems of national importance. His work has included leadership and technical roles involving the Department of Defense, DARPA, the Defense Logistics Agency, and other government organizations. He has focused on how intelligent systems can support complex decision making, coordination, logistics, engineering, and organizational effectiveness.

Michael DeBellis has worked in artificial intelligence, knowledge representation, software engineering, and enterprise systems for more than three decades. His experience includes work on knowledge-based software engineering for the U.S. Air Force, technical leadership in enterprise consulting, and service as a technical partner at Deloitte Consulting. In recent years, he has worked as an independent consultant focused on Retrieval-Augmented Generation, Semantic Web technologies, ontologies, knowledge graphs, and the integration of Large Language Models with enterprise data. His book, Designing Semantic Knowledge Graphs: A Guide to Utilizing Semantic Web Technologies for Agile Data Management, is scheduled for publication in November.

What We Do

Modern engineering and enterprise organizations already contain enormous amounts of knowledge about how to do work well. That knowledge is found in standards, checklists, best practices, requirements, architectures, test plans, review comments, issue trackers, project schedules, risk registers, and the minds of experienced practitioners. The challenge is that this knowledge is often fragmented across tools, documents, teams, contractors, and communication channels.

We develop concepts, architectures, and prototype systems that help make this knowledge explicit, connected, and actionable.

Our current focus is the integration of Large Language Models, AI agents, Retrieval-Augmented Generation, and Semantic Knowledge Graphs. We use standards-based semantic technologies such as RDF, OWL, SHACL, and SPARQL to represent the meaning, structure, provenance, and validation rules associated with complex knowledge domains. We combine these technologies with LLM-based assistants that can help users find, interpret, validate, and act on relevant information.

This approach is especially relevant for systems engineering and other high-consequence domains where AI tools must be grounded, explainable, auditable, and under human control. Rather than treating AI as an unconstrained decision maker, we focus on AI assistants that operate over explicit, governed, traceable knowledge.

Current Areas of Interest

Semantic knowledge graphs for systems engineering and enterprise data management
AI assistants for requirements analysis, trade studies, decision support, and project coordination
Retrieval-Augmented Generation systems grounded in curated enterprise knowledge
Integration of LLMs with ontologies, graph databases, validation rules, and provenance models
Tools that help teams manage change impact, traceability, verification evidence, and governance obligations
Human-centered AI systems that augment expert judgment rather than replace it

Our Perspective

The next generation of AI-enabled engineering tools should do more than summarize documents or answer isolated questions. They should help teams connect requirements, architectures, interfaces, tests, risks, decisions, assumptions, schedules, and evidence across the lifecycle of a project.

We believe the most useful systems will combine the strengths of neural AI and symbolic knowledge representation. LLMs provide powerful language understanding and interaction capabilities. Semantic Knowledge Graphs provide explicit meaning, validation, provenance, traceability, and auditable reasoning. Together, they can support trustworthy AI assistants for complex engineering and organizational work.

Website Status

This public website is currently being updated as Integrated Information User Support Environments develops its current focus and public materials. Additional information about our work, publications, and prototype systems will be added over time.

For current professional background, please see:

Robert Neches — Bio / CV
Michael DeBellis — Bio / CV

For inquiries, contact: [mdebellissf@gmail.com]