Learning Comes After Action
An organization learns from a choice when it can compare what it expected with what actually happened.
Read the noteInsights
Notes from inside the work: data and analytics modernization, enterprise architecture, semantic foundations, AI governance, and the questions that become Decision Intelligence.
I write about the moments when a technical choice meets a human responsibility. Drafts remain private until I am ready to put them into the world.
Published elsewhere
Explore Articles Looking for external long-form essays and publication series? Explore Articles on Medium and beyond.
Four-note sequence / 02
Notes on the questions behind better decisions
A short sequence from the work: why data needs context, why decisions need memory, and how organizations learn after they act.
Next questionWhat would change if the data, the decision and the outcome could teach one another?
A note to begin / 001The
decision
gap.
decision
gap.
Every Decision Leaves a Trace
We are usually good at recording what happened after a choice. We are less consistent at recording what was known, assumed, and rejected before it.
Read the noteAI Needs an Accountable Place
A recommendation can help with a decision only when the boundary between model contribution and human judgment is clear.
Read the noteData Is Not Context
A signal becomes useful at the moment of choice only when its meaning, objective, timing, and constraints travel with it.
Read the noteWhy Enterprise Systems Remember Transactions but Forget Decisions
We are often able to reconstruct what happened financially. We are much less able to explain what people believed, rejected, or changed before it happened.
Read the noteMigration Is Not Modernization
Moving a workload changes where technology runs. Modernization changes what an organization can know, decide, and improve.
Read the noteThe AI Learns From Data. Who Learns From the Decision?
I keep coming back to a simple question: if a model can learn from feedback, how does the organization learn from the choices it makes?
Read the noteFrom Semantic Layer to Decision Context
A semantic layer gives data a common meaning. Decision context gives that meaning a purpose, a boundary, and a consequence.
Read the noteWhen AI Recommends and Humans Decide
The important boundary sits between a recommendation, the responsibility for acting on it, and the quality of the handoff.
Read the noteWhy AI Agents Need More Than Access to Enterprise Data
Giving an agent access to data is only the beginning. It still needs to understand what matters, what is allowed, and when a person should take over.
Read the noteThis notebook is connected to the private editor. I will keep adding notes as the questions become clear and the wording is ready. Open the owner editor