Field Notes
Writing on world models, agents, causal inference, and the future of operational decision-making.
Should we commit $400M more to memory chip inventory?
Memory chips are being discontinued and prices are up tenfold. One decision, worked end to end: the cascade across five functions, the console, every move scored side by side, and the recommendation with its trace.
Ergodic Recognised in Nucleus Research's 2026 Control Tower Technology Value Matrix
Nucleus Research has named Ergodic in its 2026 Control Tower Technology Value Matrix. Here's how we're working in supply chain.
Ergodic Wins Innovate UK Backing for Causal World Models
Over the next six months, our team will research causal world models that can learn a task in one environment and perform it reliably in another.
Context Graphs: Seeing Your Supply Chain As It Actually Is
Your supply chain isn't a list of vendors. It's a living graph of relationships, flows, and dependencies. Here's why that distinction matters when things go wrong.
The Causal Learning Series: An Introduction
A guided, ground-up path through causality, the discipline behind every world model. Start with why correlation was never enough, and build toward models that reason about cause and effect.
Agent Capabilities Are Growing, But The Architecture Hasn't Caught Up.
Agents are being handed decisions of increasing consequence, but the frameworks underneath were built for tasks. The missing layer is a world model that lets an agent simulate before it acts.
The Revenue Decision Agent: Simulate the Quarter Before You Commit
Revenue tools score deals and transcribe calls. None of them can answer the question every CRO carries into the quarter: what happens if we act?
The Operations Decision Agent: Run the What-if Before You Commit Inventory
Planning tools optimise a fixed model of the operation, and dashboards report problems after they've happened. Neither can answer the question that defines operations work.
Loops: How AI Agents Get Things Done, and How a World Model Rebuilds the Loop
Every AI agent runs on a loop, repeated until a goal is met. A world model inserts a step that has never been there, and wraps a slower loop around the whole thing that learns from what happened.
AI Agents Need A Robust Data Foundation
Before AI agents can optimise your supply chain, they need something to reason about. That something is a context graph they can traverse, query, and simulate against.
What Are Enterprise World Models?
A world model carries a causal representation of a system and evolves it forward under the actions you take. Applied to an enterprise, that means state, transitions, and an entity graph carrying the constraints the business obeys.
The Causal Learning Series lives in the Ergodic documentation.
Go from reading to running.
Bring a decision. We'll discuss how to model it and what data a simulation would need.
Talk To An Expert