A unified control plane for every connected device in your building. Humans and AI agents are both first-class users.
Embrava Connect Cloud (ECC) is an AI-native device management platform built for the workplace. It handles identity, state, provisioning, updates, and coordination for every device you deploy — from small screens to sensors to signage to whatever you plug in next. Everything else is what you do with it.
What makes ECC different: AI agents aren't an add-on. They're users. An agent authenticates with OAuth2 client credentials, receives role-based permissions, and interacts with every device and workspace through the same API surface as a human operator. The platform doesn't distinguish between a facilities manager booking a room and an AI agent optimizing floor utilization at 2am.
AI agents authenticate, hold roles, and operate devices through the same APIs as your team. Not a bolt-on. The architecture.
A single device can run an embedded today and a different one tomorrow. No swap, no remake. No new vendor.
From bare-metal firmware to Android appliances to enterprise MDM-managed hardware — all managed through the same control plane.
20+ native integrations plus an open API and MCP server mean the platform speaks the language of your business and your AI tools.
Add capacity, add modules, add device types, add agents — without re-architecting anything.
Multi-tenant isolation, SSO, audit logging, agent RBAC, and the security posture IT teams expect.
ECC exposes a full MCP (Model Context Protocol) server so any AI agent can discover, authenticate, and operate your workplace infrastructure. Agents get their own identity, their own RBAC roles, and their own audit trail. Create an agent, assign it permissions, hand it a client secret, and it's operational.
Standard A2A (Agent-to-Agent) discovery via /.well-known/agent-card.json means any agent framework can find ECC and understand what it can do. No custom integration. No vendor lock-in. Just the open protocols.
ECC doesn't just show you data and wait for you to act. The platform includes a policy engine and workspace intelligence layer that lets agents make decisions and take action in real time.
Define rules: if a desk is empty for 15 minutes, release it. If occupancy drops below 18% after 2pm, consolidate zones. The platform enforces them automatically.
Occupancy prediction from booking history and device telemetry. Anomaly detection. Booking-vs-occupancy gap analyses. Pattern recognition by day, time, and team.
Upload an architectural floor plan. Claude Vision detects every room, labels it, and maps it to the platform. No manual drawing. Minutes, not days.
Firmware vulnerability scanning, config rollouts, auto-rollback. Your devices stay current without a human in the loop.
Employee offboarded in HRIS? Bookings revoked, licenses updated, devices reconfigured. Automatically.
Event tomorrow? Signage updated, rooms reserved, wayfinding configured, resources allocated. One trigger, everything moves.
The platform decides what that endpoint does. A module — desk booking, room scheduling, signage, alerts — is just software running on the platform. Swap the module and the device takes on a new job. That's the whole idea, and it's why a customer who buys one product today is really investing in every product we'll ever build.
The same is true for agents. An AI agent that books desks today can optimize signage tomorrow and manage fleet health next quarter. Same credentials, same permissions model, same audit trail. The platform grows. The agents grow with it.
Sixty minutes at our Hudson Yards Experience Center — every product, the engineers, your stack pre-wired. Or just send us a message.