Growing the Symbolic AI Community in the Age of Generative AI
Generative AI has real limits, and symbolic AI covers many of them. This piece introduced the symbolic toolkit, showed where it complements probabilistic methods, and made the case for joining a community of practice to speed your own learning while lowering the barrier for everyone else.

The Epistemic Loop: A Model for Hybrid AI
A blueprint for systems that pair symbolic and probabilistic AI so each one's weaknesses are covered by the other, with humans originating and validating what the system produces.

Metaxy: Perfecting the Art of Doing Nothing
A look at Metaxy, an open-source project that cuts the cost of running multimodal data systems in production. By isolating exactly what changed and reprocessing only that, pipelines free up budget for more experiments.

Integrating AI in Medical School Education: Connecticut Edition
A survey of how Connecticut medical schools are folding AI into their curricula, from interactive simulations to hands-on exercises that push students toward sharper clinical thinking.

Large Language Models as Generative Ontologists
Where LLMs help with ontology work and where they fall down. The piece lays out how knowledge is structured and layered, then what that means for anyone using LLMs to support knowledge management and ontological engineering.

AI Systems Thinking: What Leaders Need to Know Before the Demo
Most prototypes fail on the way to enterprise-grade because they were never grounded in a verified problem. Demos also fail when sponsors and the wider organization want different things and nobody names the gap.

Building Knowledge Graphs with an AI Wingman
Three SPARQL queries of increasing complexity, showing where generative AI genuinely assists knowledge managers and ontologists with the human still at the center. Covers lexical enrichment, controlled semantic enrichment, and cross-graph reconciliation.

The Prototype Trap: Why AI Governance Failures Often Reflect Institutions
A look behind the curtain at governance and stewardship around prototypes, and why organizations stall. Usually the stated values and the behaviors that actually get rewarded are pulling in opposite directions.

From Vibe Coding to Product Building
Vibe coding opened the door to rapid prototyping and then led straight back to requirements engineering and documentation. The piece argues it is an early stage in a maturing practice of capturing tacit knowledge.

The Skill Stack: Turning Judgment into Reusable AI Infrastructure
Ad hoc prompting behaves a lot like throwaway scripting. This one covers how to move from disposable prompts to a reusable, extensible system that holds a team's collective knowledge and skills.