Articles

Five episodes ago, we started with a simple problem: AI agents crashing production systems because nobody had thought about isolation. This is the final episode. Here is how all the pieces fit together, and why the architecture matters more than any single model you will ever rent. This is the final episode of Pragmatic AI […]

The market is currently flooded with certified AI experts who know everything about language models and absolutely nothing about your supply chain. The hard truth is that the best person to implement AI in your company isn’t a developer. It’s your current consultant. The AI Expert Bubble Over the past year, a new industry has […]

Another week, another simulation. This time, picture an AI agent that was tasked with analyzing inventory. It ran an unoptimized search loop on the live database, locking tables and halting the shipping bay for two hours. Connecting probabilistic models to your production systems looks nice to me only on a slide. Overall, it’s a gamble […]

A few western companies trained their models on the collective data of the internet. Now they sell that intelligence back to us through metered API calls with rising token costs and strict operational limits. I described this situation more and more often in recent months, since from around April 2026 the infinite token quota plans […]

Let’s simulate that your AI agent recommended ordering a replacement valve based on a semantic search of a maintenance manual. It missed the fact that the valve is part of a retired line. A €15,000 inventory mistake. AI hallucinations in real world environments can cost a lot. This is the fourth episode of Pragmatic AI […]

Back in April, DeepSeek released a model that rivals semi-frontier models basically for free. Your ERP vendor is probably charging you extra to use some last year’s version behind a paywall. To me, something in this equation does not add up. Training a Large Language Model is an extraordinarily expensive business. The capital required for […]

Imagine to be part of an integration team that just spent three months and €50,000 building custom APIs so an AI agent could read inventory and track orders. The next day, the system vendor updated their schema, and the entire stack crashes. Integrating AI with enterprise systems is turning into a nightmare of custom development […]

We are told that generative AI is a clean, weightless cloud of pure intelligence. Down on earth, the reality is far more material: it is an extractive empire of cheap labor, diverted rivers, and burnt grids. Reading Karen Hao’s book, Empire of AI, is a sobering experience for anyone working in technology. I advice any […]

Picture this: your logistics team just deployed an AI agent to fix stock discrepancies. It ran for fourteen minutes before anyone checked. By that time, it had already modified three purchase orders and locked a database table. Understanding the difference between a language model and an autonomous agent determines whether your AI is a tool […]

Every executive planning session is probably chasing the same dream: an ERP that behaves like Lego bricks. Today, the hype around vibe coding promises to let anyone snap these bricks together with zero engineering experience. But there is a vast gap between generating a personal app on your laptop and running an enterprise workflow. And […]

Most enterprise AI failures don’t start with bad code or dirty databases. They start with a single, unasked question during the initial steering committee: do we need a deterministic solution, or a probabilistic one? Lately, I’ve been reading a lot of posts and articles related to this topic, and all of them where very focused […]

s your executive steering committee pushing hard for generative AI adoption? Well, your employees are actively delivering it. They are doing so by purchasing personal Claude subscriptions with their own credit cards and using them to process your company’s data. A quiet, undocumented migration of corporate intellectual property into public models, happening in plain sight. […]