Code ParticleCODE PARTICLE

Free field guide

A Working Demo Is Not a Working System

A short field guide for leaders adopting AI. A working demo proves capability. Before the business can rely on it, you still have to answer what it can touch, how you know it is right, and who is responsible when it is wrong.

Download the guideFree PDF. No signup.
Cover of the field guide, A Working Demo Is Not a Working System, by Nicola Onassis of Code Particle

What is inside

  • Why an impressive demo can still stall before it reaches real work.
  • The three questions to ask before you trust any AI output: what it can touch, how you know it is right, who is accountable when it is wrong.
  • A concrete example of a flawless-looking AI result that quietly fails all three.
  • What "ready" actually looks like, and why careful is not the same as slow.
  • Where to start so a mistake is cheap, and a full self-assessment to run on your own.

A working demo proves capability. The business still has to prove control.

Who it is for

For owners and executives in law, accounting, healthcare, finance, and other regulated businesses who own the AI decision but do not have an engineering team. If you have seen a strong demo and then watched the project stall before real use, this guide was written for you.

Written by Nicola Onassis, founder of Code Particle and host of the podcast AI Decisions You Have to Defend. Code Particle has built software for regulated, high-stakes industries since 2008.

Have an AI use case you are weighing?

Read the guide first. If it raises a question about a real project, bring that use case to us for a free 30-minute conversation. We will look at the practical issues with you: access, verification, responsibility, and whether it is a sensible place to start.

Bring us your AI use case Download the guide