My IT Blog

Running an AI model is one thing. Running an AI agent that can actually perform actions inside an enterprise environment is a very different architecture problem. A traditional model interaction is roughly: Prompt → Model → Response An agent can go much further: Objective → Reason → Retrieve → Select Tool → Execute → Observe For me, the important word is execute . Once an AI system can call APIs and change something, this is no longer only a model-hosting problem. It becomes a security, infrastructure and operational architecture problem. Think of an Agent as a Workload with Authority Imagine an infrastructure operations agent. It might read monitoring data, inspect logs, search documentation and recommend a remediation. So far, the blast radius is limited. But then somebody asks: "Why can't it just fix the problem automatically?" That question changes the architecture. An AI agent is not only an intelligent application. It is an appl...

Private AI on VMware Cloud Foundation: How I Would Architect It There is a big difference between running an AI model and building an AI platform. We can attach GPUs to infrastructure, deploy an LLM and expose an endpoint fairly quickly. That proves the technology works. It does not mean we have built an enterprise AI platform. Once different application teams start asking for models, GPUs, knowledge bases and AI agents, the architecture problem changes. Now we need to think about shared infrastructure, isolation, model lifecycle, data access, GPU utilization, security, observability and cost. That is the point where I think VMware Cloud Foundation becomes more interesting. In this article, I want to look at the problem from an architect's perspective: If I already have VMware Cloud Foundation, how would I structure it to provide Private AI as a platform rather than simply running a few AI workloads? Start with the Consumer, Not the GPU One mistake I would av...