THE ENCLO PLATFORM · EDGE TO CLOUD

Run your AI at every site.
One platform, edge to cloud.

Enclo deploys and updates your models across every customer site: on edge devices next to the cameras, robots, scanners and sensors, on GPU clusters in the datacenter, and in your cloud. Every release is checked, tested on live data and rolled out safely, site by site. It runs in your environment, and your data stays there.

Get the technical overview for your security and platform team.

Sample data

Overview, 14 sites

3 on v2.4.1, 11 on v2.4.0, 1 held

  • Plant 2v2.4.1
  • Lab 3v2.4.1
  • Hospital Av2.4.1
  • Warehouse 1v2.4.0
  • Robot cell 4v2.4.0
  • Clinic Bv2.4.0
  • Site 07held v2.4.0
  • Plant 5v2.4.0
  • Depot 2v2.4.0
  • Lab 1v2.4.0
  • Line 3v2.4.0
  • Clinic Dv2.4.0
  • Yard 1v2.4.0
  • Site 12v2.4.0

Release

v2.4.1rolling

3 of 14 sites, canary passed

Activity

Site 07 held, reverted to v2.4.0, reason: gate failed

The other 13 sites continue.

Fig. 1. The console with sample data: fourteen sites across plants, labs, hospitals and warehouses, a signed release rolling out after the canary passed, and one site held and reverted on its own while the others continue.

Open the console preview

Every site

One view

Which model runs where, on every device at every customer site

Safe

Releases

Signed, capacity-checked, shadow-tested, rolled out in stages with automatic rollback

Edge → cloud

One system

Edge devices, on-prem GPU clusters and your cloud accounts, on one provider or several, together

Yours

Environment

Runs at your sites, in your datacenter and cloud. Your data stays there

The shape of it

Inference at the edge. Heavy compute behind it. One platform across all of it.

ONE PLATFORM · RUNS IN YOUR ENVIRONMENT

THE EDGE

Where data is born

  • Cameras
  • Robots
  • Scanners & sensors
inference

EDGE INFERENCE

Vision runs at the edge

  • Jetson-class
  • On-prem GPU box
  • Edge server
heavy workmodels · results

YOUR GPU COMPUTE

On-prem or your cloud

  • Heavy models
  • Training
  • Multi-site rollout
Fig. 2. Data is born at the edge and vision runs there, on an edge box next to the cameras, robots, scanners and sensors. Heavier work goes to your GPU compute, on-prem or in your cloud, and models and results come back. All of it runs inside your environment.

The gap

Shipping a model once is easy. Shipping every version to every site isn't.

Physical AI runs where the data is born: in operating rooms, labs, plants and warehouses. The first customer site works because your best engineers sat with it. Then the second site has different hardware, different network rules and a different version of your model, and nobody can say for sure which version runs where. Every new site gets slower, not faster.

What Enclo does

One platform between your models and every site.

The Enclo platform connects every site to your GPU compute and your model registry. Each new site is brought up the same way, and each new model version goes through the same checks before it reaches a device. We install it in your environment and train your team, and your team runs it from there.

Fig. 3. Today each site drifts to its own model version and one stops reporting. With Enclo every site runs the one signed release, and the one held device is visible instead of hidden.

Why us

Built on experience running this at hyperscale.

Enclo's engineering background: multi-tenant GPU inference platforms at a major cloud provider, Kubernetes-based GPU orchestration and fractional-GPU scheduling, and installers shipped into air-gapped and restricted networks. The same discipline, applied to your footprint.

Multi-tenant GPU inference at cloud scaleKubernetes GPU orchestrationEdge + hybrid deploymentAir-gapped delivery experience

The platform

Every model, every device, every site. One console.

See what runs where, ship a new model version safely, and roll it back in one step if it isn't ready for a site.

Every model, every site

Which version runs on which device, at every customer site, in one view. Drift shows up before a customer notices it.

Signed releases

Each release is signed in your registry and verified on the device before it installs.

Capacity check

A new model only goes where the GPU has room for it. Devices that don't fit stay on the current version.

Shadow test on live data

The new model runs beside the current one on each site's own data, and goes live only when it clears the bar your team sets.

Staged rollout, automatic rollback

One device first, then wave by wave. If a release fails anywhere, every device goes back on its own.

Edge to cloud, one system

Edge devices, on-prem GPU clusters and your AWS, Azure or GCP accounts, with data rules set per site. Sites keep running when the network drops.

Ways to deploy

One platform. Three ways to run it.

Platform details

Edge + cloud

Vision and custom models running next to the cameras, robots, scanners and sensors, with the heavier models in your datacenter or cloud. The platform keeps every site on the right release.

On-prem & air-gapped

The full platform inside a datacenter or colo, from bare metal up, for data that can't leave the building. Runs fully air-gapped when the environment requires it.

Multi-tenant

For products that run other people's workloads: tenant isolation across compute, network and data, on shared GPU capacity, with each customer kept separate.

Where it runs

At the edge, on-prem, and in your cloud.

The line that matters isn't on-prem versus cloud. It's whose account it runs in. The Enclo platform runs in environments you control: your customer sites, your datacenter and your cloud accounts. Those accounts can sit on different providers, so a startup with AWS and Azure credits uses both, and a customer who standardized on Google Cloud is one more site on the same release.

Server rack in a datacenterA rack elevation drawn in outline with four installed GPU servers, each with status lights, and two empty slots: the customer's own datacenter or colo.

Your datacenter or colo

Bare metal to production Kubernetes: GPU drivers, scheduling, storage, serving. Fully air-gapped when your environment requires it.

Edge inference deviceA small fanless edge box with a status light, drawn in outline. Its uplink to the core drops for part of the loop and the light keeps breathing: the device keeps serving when the connection is down.

The edge

Inference at the edge, where your data is born: edge boxes and on-prem GPUs next to the cameras, robots, scanners, and sensors. Heavier work is handed back to your datacenter or cloud. Keeps serving when the network drops.

GPU instances inside your own cloud accountThe same GPU server drawn inside a dashed boundary labelled your account: the hardware is the cloud provider's, the account around it is yours.YOUR ACCOUNT

Your cloud accounts

GPU workloads inside your own AWS, Azure, GCP or OCI accounts, and more than one at a time. Spend AWS and Azure credits on the same models, or run a customer on the cloud they mandate. You pay each provider directly.

Getting started

Assess. Install. Run and grow.

Scoped and priced before anything starts, with no open-ended meter.

  1. 01

    Assess

    A short, fixed-fee review of your sites, devices, models and data rules. You get an architecture, GPU sizing and a plan for bringing the platform up, with a price.

  2. 02

    Install

    We install the Enclo platform in your environment, connect your devices, models and GPU compute, and prove it against acceptance criteria agreed upfront.

  3. 03

    Run and grow

    Your team runs it, with runbooks and training. A yearly platform license covers new platform releases and direct engineering support. Every new site goes through the same platform.

Commitments

What you can count on.

Your team stays in control

The platform is built so your own engineers can run it, with runbooks, training and infrastructure-as-code in your repositories. No lock-in.

Your environment, your data

Enclo runs at your customer sites, in your datacenter and in your cloud accounts, whichever providers they are on. Your data stays in your environment, and we only get the access you grant.

Your models stay yours

The models, agents and business logic are yours. Enclo is the platform they run and ship on, and we never compete with the teams we support.

See the platform on your sites.

A 30-minute call with an engineer, not a sales rep. Bring your sites, devices and models. We'll walk you through the platform and give you a straight answer on whether it fits.