About
Most AI tells you what. INQ tells you why.
Inquisition AI was built by engineers — people who have stood in front of a failed unit, an off-spec batch, or a trip report and been accountable for the answer. The problems worth bringing to a machine — a failure you can't explain, a decision you can't undo, a result that doesn't add up — deserve more than the first thing that sounds right. So we built the tool we wanted on those days.
Why it exists
The stanceEngineering already has a discipline for hard failures: structured root cause analysis. Lay out the possible causes, demand the evidence that would confirm or eliminate each one, rule things out on facts, and don't call it solved until the answer survives scrutiny. It works — but done by hand it's slow, and most AI does the opposite: autocomplete a plausible paragraph and present it with confidence.
INQ puts that discipline into the machine. Instead of racing to an answer, it runs the investigation: it maps the possibilities, asks what would prove or disprove each one, and holds the question open until the evidence closes it. The method comes from engineering — but nothing about it is exclusive to engineers. The same rigor works on a clinical finding, a business result, or any outcome you can describe.
Principles
How we build itEvidence over assertion
A confident paragraph isn't a conclusion. INQ asks for the evidence that would confirm or eliminate each possibility, and shows its work so you can check it.
Structure over one-shot
Hard problems don't yield to a single prompt. INQ holds the problem open — mapping causes, weighing branches, ruling things out — until the answer actually holds up.
Your reasoning, extended
It doesn't replace your judgment; it stress-tests it. You stay in the loop, answer the questions, and leave with something you can defend to anyone.
Who it's for
By engineers · not only for themFirst, our own people: reliability and maintenance engineers tracing equipment failures, process engineers chasing off-spec product, operators working incidents. That's who INQ was built for, and it shows in the tool — evidence strength ratings, ruled-out causes kept on the record, reports you can hand to a plant manager. But accountability for being right isn't an engineering monopoly: clinicians explaining anomalous findings, analysts defending a conclusion, anyone doing due diligence where a wrong answer has consequences — if you can describe the outcome, INQ can help you trace it.
See how it works.
Walk through a real investigation, or start your own — the first one is free.