Test HyperAI on your real workload. In 14 days. In your country.
A sovereign GPU cluster, sized for your use case, deployed in Cairo or Riyadh, on the MomentumX sovereign cloud platform. Bring your training data, your model weights, or your inference workload. We’ll size the cluster, do the onboarding, and give you 14 days of capacity to prove it.
What’s in the POC
Your choice of H100, H200, or A100 SKU based on workload sizing.
Cairo, Riyadh, or your facility. Your data stays inside the country you select.
Environment setup, NVIDIA stack configuration, model loading, performance tuning — done with a MomentumX engineer.
Open architecture, your choice of training and inference runtime, container orchestration, and high-performance interconnect — we’ll match your team’s existing toolchain.
For a real workload — a training run, an inference benchmark, or an end‑to‑end pipeline test.
Your data and models leave with you, no obligation, no proprietary formats.
Apply for the POC
Tell us your model size, your data classification, and your timeline. A MomentumX architect responds within 48 hours with a sized cluster and a deployment date.
No commitment. NDA on request before you share workload details.
Who the POC is for
Filter, don’t sell. We’ll be honest if it’s not a fit.
KSA Financial Institutions
SAMA‑regulated banks, insurance, fintechs running fraud detection, document understanding, customer analytics, or regulatory reporting AI. The POC unlocks sovereign LLM deployment with a documented compliance posture.
Telcos and Government‑adjacent
Egyptian and KSA telcos, energy, utilities, defence‑adjacent. Workloads where the data is national‑interest sensitive and a hyperscaler region is not a sufficient answer.
Egyptian Enterprises with PDPL Scope
Banks, healthcare providers, insurance, government‑linked entities running AI on personal data covered by the Egypt PDPL.
How the 14 days run
From application to decision point.
| Day | What happens |
|---|---|
| Day 0 | You submit the form. We respond within 48 hours with a sized configuration and a deployment date. |
| Day 1 | Kickoff call with a MomentumX engineer. Define the workload. Confirm cluster size and SKU. |
| Day 2–3 | Cluster provisioned in Cairo or Riyadh. Environment, isolated tenant, container namespace, model storage. Onboarding session. |
| Day 4–11 | Your team runs the workload. We’re available for support. Daily check‑ins if you want them. |
| Day 12 | Benchmarks captured. TCO comparison vs your current GPU spend documented. |
| Day 13–14 | Decision point — sign a deployment contract, take the data and exit, or extend on a month‑to‑month basis. |
The POC is free. The decision afterward is yours.
We don’t charge for the 14‑day POC. We invest the GPU capacity and the engineering hours because the conversion rate from “ran a real workload on HyperAI” to “signed a deployment contract” is high enough to justify it. If you exit at the end of the 14 days with no obligation, that’s fine — the POC was the cost of finding out.
Frequently asked
Will the POC actually be on production‑grade hardware?
Yes. The POC cluster runs on the same HyperEdge 500 + GPU configuration as a deployment cluster. Same NVLink topology, same networking, same software stack.
Can we run benchmarks against our current setup?
Yes — we encourage it. Bring your existing benchmark suite. We’ll help you set up a controlled comparison.
Do you require an NDA before the POC?
We’ll sign a mutual NDA on Day 0. Standard terms.
Does our data stay in country during the POC?
Yes. The cluster is provisioned inside Egypt or KSA. Your data, models, and outputs never leave the jurisdiction you select. Documented in the POC engagement letter.
Can we extend the POC?
Yes. After Day 14, you can extend on a month‑to‑month basis at the standard cluster rate while you finalise procurement.
What happens if we don’t sign a deployment contract?
We wipe the cluster, you take your data and models, no obligation. We may follow up after 30 days to ask what we could have done differently — that’s it.
Apply for the POC
Tell us about your workload. An architect responds within 48 hours with a sized configuration and a deployment date.
