Perspectives
I write about how a company can preserve its knowledge and use AI to make better decisions. Energy, investment and my work at TheryOS are the starting point for these ideas.
Join the conversation
Receive my new articles on world models, artificial intelligence and business decisions, informed by my experience in energy and finance.
Featured Article
Why we decided to bring TheryOS into investment
Why we are bringing TheryOS world models to investment, and what we hope to achieve with Thery Dynamic and Thery Signals: better decisions with traceability.
All Articles
Vertical AGI: why I work with sector-specific world models
Energy taught me that knowing a fact and knowing what to do with it are different problems. That distinction is at the heart of my work on decision systems.
How to evaluate an energy AI: what a test actually proves
Repeating a calculation, respecting a permission and lowering a cost are different achievements. An evaluation should say which one it tested and under what conditions.
From an energy bill to a decision: the savings that matter
An option can lower the electricity bill while increasing the company’s costs. This example shows why reading a bill properly is only the beginning.
A company’s knowledge should survive a change of AI
Changing AI providers should not mean losing the reasons, agreements and relationships a company has built. Here is how to check that this knowledge remains usable.
A company needs continuity to work with AI
Business tasks span waiting periods, people and tools. Continuity makes it possible to resume progress and see what remains to be done without rebuilding the work in every conversation.
Every decision leaves knowledge: how a company learns
An improvement after a decision can have several causes. I explain how to preserve the expectation, understand what was actually done and turn experience into a useful criterion for the next case.
From an LLM with tools to sector intelligence: what remains to be proven
Tools expand what an LLM can do. The next challenge is understanding the business objective, adapting to new problems and working with the autonomy the company grants it.
Context is not memory: why business AI needs an authorized reality
An assistant may remember an approval that is no longer valid. Using examples involving contracts, price changes and responsibilities, I distinguish memory from context inside a company.
Advanced RAG is not a world model
RAG helps find and connect documents. A decision requires linking that information to contracts, facilities and consequences. I explain the difference through a tariff comparison.
Towards a Deterministic Vertical AGI for Energy Systems
A study of how to represent the energy sector in a world model and use its data and rules to make decisions. It explains how to follow the calculation and review the result.
From Integrated Energy Management to Vertical AGI: The First Patent
Patent WO2024/194517 A1 brings together information from meters, bills and installations to understand energy use. I explain how that idea connects with my work at TheryOS.