How it works in practice

How do you test an AI system before it touches real customers?

We run the system in shadow mode first: it makes real decisions against real data, but a person reviews them before anything reaches a customer, including email sends, order actions, or messages. Only once it holds up across edge cases and failure scenarios does it get permission to act on its own, with a human escalation path left in place.

By Precipitate · Updated 14 August 2026

Want this answered for your own business?

Get a straight answer

Testing starts before any code runs against a live situation. We map the manual process the system is replacing first, so we know what a correct decision actually looks like before we can grade the system against it. Then we build guardrails around it: which tools it's allowed to touch on its own, and which actions still need a person to approve before they go out.

Once it's built, it runs in shadow mode against real or recent data. It makes its decisions and would take its actions, but instead of sending the email or placing the order, the action gets logged instead of executed. We push it through cases meant to break it: a tool that times out, input it wasn't expecting, a decision that's genuinely ambiguous. Then we watch what happens: does it retry sensibly, escalate to a person, or fail without telling anyone. Silent failure is disqualifying.

Go-live is gradual, not a switch flip. We start it on a narrow slice, one channel or one type of decision, and widen its scope once it's run clean for a while. Because we operate the systems we build instead of handing over code and leaving, monitoring doesn't stop at launch: we keep watching the logs, and anything the system isn't confident about keeps routing to a person rather than getting guessed at. We run this same process on our own systems too. Precipitate currently operates 88 systems and more than 110 scheduled jobs across 40-plus integrations, and each one went through a dry run before it touched anything live.

Related questions

Do you test with real customer data before launch?

Where possible, yes: shadow mode runs against real or recent data so the test reflects what the system will actually see. For situations that haven't happened yet, like a rare complaint or an unusual order, we write synthetic cases so the system gets tested against them before a real customer becomes the first example.

What happens if the system makes a mistake after it goes live?

It's built to escalate rather than guess: anything outside what it's confident handling gets flagged to a person instead of forced through. Because Precipitate operates the systems it builds, someone is watching the logs after launch too, not just during testing.

Wondering what a system like this would own in your business? Tell us what the manual work is, and we will tell you honestly what a machine can take off your plate and what still needs a person.

One reply from a person, usually same day. No deck, no discovery call, no sales sequence.