Articles

Thinking behind the systems.

Practical perspectives on agent design, trustworthy outputs and the choices behind useful Data and AI workflows.

Research and practical notes on Data, AI and agent design.

01

Agent Design / 9 min read / Published

Dashboard, automation or AI agent? Choosing the right system for the workflow

A practical framework for deciding whether a workflow needs better information, fixed automation, AI assistance or bounded autonomy.

Read article
02

Agent Evaluation / 9 min read / Published

How do you know an AI agent is ready? A practical evaluation framework

A practical framework for deciding whether an agent has produced enough evidence to enter a real workflow.

Read article
03

Human Oversight / 9 min read / Published

Designing AI agents people can review and trust

How to make evidence, uncertainty, actions and failures visible enough for meaningful human oversight.

Read article