FAQ

Questions we get asked

The ones that come up before a deployment — about where the data sits, what an agent is allowed to do, and how to start small.

Do we have to replace our ERP or CRM?

No. DYNARA sits on top of the systems you already run. Your team keeps using the ERP as before; what changes is that they can ask it questions in plain language and have an agent do the data entry. Existing processes stay as they are.

Where does our data live? Can we self-host?

Two options: hosted on infrastructure we operate, or deployed inside your own environment so that nothing leaves your network. Private deployment is scoped per project.

Is our data used to train models?

No. The underlying models are used under enterprise agreements that exclude customer input from training. A written summary can be provided for your security review.

How are permissions handled?

Answers are limited to what the person asking is already allowed to see. Permissions are not a layer bolted on for AI: agents read through the same access rules as the rest of the platform, so an agent cannot surface something its user could not have opened.

What stops an agent from inventing an answer, or taking a wrong action?

Three things. Answers come only from your own knowledge sources, each with a citation you can open. When nothing matches, the agent says so rather than guessing. And actions that commit something — submitting, paying, deleting — are gated behind human confirmation.

Can agents call our APIs and write back into our systems?

Yes. Agents use tools: internal APIs, platform data tables and workflows. What each agent may call is configured per agent rather than granted wholesale, so a reader and a writer can be two different agents.

How is this different from ChatGPT or Copilot?

Those are good at general reasoning but do not know your catalogue, your stock or your contracts, and they do not file anything into your systems. DYNARA connects the model to your own knowledge, your own permission model and your own workflows.

How fast are answers?

Around ten seconds for an ordinary question; thirty to sixty seconds when the answer has to be assembled from several sources. As one measured before-and-after: the same question took 12.7 to 17.3 seconds before a routing change and 7.9 seconds after.

Can we start small?

Yes. QSyn has a free tier for trying the platform, and a first phase scoped to a single workflow is the usual way to begin before anything larger.

What happens to our data if we stop using DYNARA?

It is yours and it exports in full. The platform runs on standard cloud infrastructure, so there is no proprietary format or machine of ours you would be stranded on.

Something not answered here?

Send it over. Describe the workflow as it runs today and we will come back on whether an agent is the right tool for it.

Talk to us [email protected]