How It Works
Learn how Dexal installs private AI on your own hardware in 72 hours — no cloud, no subscriptions.
Book a Free 20-Minute Call
John reviews your workflow — what you handle, how many people use it, what's slowing you down. He recommends the right setup and gives you a clear price. No pressure, no commitment.
Runs over Zoom. You'll get a recording.
We Confirm Your Hardware
We check that your existing computer meets the requirements (most 2018+ machines with 16 GB RAM work). If you need new hardware, we recommend specific models from our tested list.
NYC metro: hand delivery available. Everywhere else: ships in 72 hours.
Hardware Configured & Shipped
We pre-configure the AI system before it ships. When it arrives, it's essentially plug-and-play. No IT department required.
Flat-rate shipping included.
Remote Activation Over Zoom
John activates and configures your system live over Zoom — typically 60–90 minutes. He connects your documents, sets up your staff accounts, and runs tests.
You watch and can ask questions in real time.
Staff Training
A 30-minute walkthrough for your team covering how to use the AI, what it's good at, and what to avoid. Everyone leaves confident.
Recording provided for future onboarding.
Live — with a Safety Net
Your AI is running. If anything doesn't work as demonstrated within 7 days, we fix it free — or refund in full. No questions asked.
7-day fix-or-refund guarantee on every installation.
Frequently Asked Questions
Do I need to be in New York?
No. All demos and activation sessions are remote via Zoom. Hardware ships nationwide in 72 hours. NYC metro clients can also opt for hand-delivery and in-person setup.
What if something breaks after the 7-day window?
Minor issues are usually fixable via Zoom in 15 minutes. Clients on the Monthly Support Retainer get priority response. Outside of retainer, John offers one-off support sessions.
Can you install on an existing computer we already own?
Yes, as long as it meets specs (typically 16 GB RAM, 50 GB free disk, Windows 10/11 or macOS). We verify this on the discovery call.
How private is it, really?
The AI model runs entirely on your hardware. No API calls to OpenAI, no data transmitted to any cloud service. We can demonstrate this during the activation call — you'll see there's no internet traffic from the AI process.