NeoEdge Data Governance and Industrial AI Orchestration Platform
NeoEdge is an innovative Data Governance and Industrial AI Orchestration Platform
By combining Edge AI, AIoT, and AI Agents, NeoEdge enables real-world Physical AI applications.
It provides users with flexible hardware configuration through software-defined hardware, empowering AI technology and creating a complete solution from edge computing to AI agents.
Quick Start
Deploy quickly and get started immediately.
Easy to manage
Simple and intuitive user interface
Cybersecurity Defense
Comprehensive security protection mechanism
Product Architecture
The NeoEdge Data Governance and Industrial AI Orchestration Platform employs a software-defined hardware architecture.
By combining containerization technology to build the Edge AI platform, and by collaborating with 7 domestic industrial computer manufacturers, we have achieved flexible and efficient collaborative management of edge devices.
NeoEdge allows for easy connection of multi-source sensor data from OT devices, enabling Sensor Fusion and data conversion via NeoFlow.
By combining this with edge AI model inference, creating a complete Physical AI loop from sensing and data fusion to intelligent decision-making.
The platform also supports uploading data to existing IT systems (MS SQL, MySQL, PostgreSQL).
Together with domestic vertical industry partners, we will jointly create various Edge AI solutions based on NeoEdge.
Caption: NeoEdge Data Governance and Industrial AI Orchestration Platform architecture diagram. Centered on the Edge AI closed loop, it delivers a complete Physical AI cycle, from data collection and model training to inference deployment and command execution.
Product core components
NeoEdge consists of three core components: NeoEdge Central, NeoEdge X, and NeoEdge Matrix.
NeoEdge Central provides SaaS services running on the public cloud and also supports a private cloud version on the ground. It offers a management interface, centralized management of edge gateways, applications, edge AI model deployment, and definition of data flow rules.
NeoEdge X is a program that runs on the gateway (IPC) and is responsible for executing the tasks defined on the gateway by NeoEdge Central. Through containerization technology, users can configure various applications, giving industrial computers different capabilities to meet the needs of different application scenarios.
NeoEdge Matrix is the data governance core of the NeoEdge platform. It integrates equipment, processes, and both real-time and historical data into a unified enterprise semantic model (Ontology), giving IT and OT data consistent context and serving as a reliable foundation for industrial AI analytics and decision-making.
NeoFlow V4
Connecting NeoEdge Matrix with AI Agent MCP's core edge intelligence services:
NeoFlow V4 combines no-code graphical logic editing with an open Custom App architecture,
enabling enterprises to quickly develop, deploy, and manage Edge AI applications.
It supports group-based dispatch to NeoEdge X industrial PCs, lowering the barrier to adoption and accelerating OT-IT integration,
while improving flexibility, maintenance efficiency, and intelligent management on the factory floor. It further connects to an MCP Server, allowing AI Agents to seamlessly link with Physical AI.
No-code graphical interface
Drag and drop to arrange, and you'll have everything at your fingertips, bidding farewell to cumbersome development processes and terminal commands.
Scalable open architecture
Provides SDKs (Python, Go) for developing custom apps and supports custom nodes.
100% Containerization Support
It is compatible with mainstream AI inference frameworks such as NVIDIA TensorRT and Intel OpenVINO.
Six major features
OT data ETL
Extract
Supports protocols including Modbus, OPC UA, MELSEC, RTSP, MQTT, and RESTful API to integrate data from heterogeneous devices
Transform
Standardizes and structures IT/OT data to improve data consistency and usability
Load
Imports data into designated IT systems, including public cloud, private cloud, MES, ERP, and databases
NeoEdge X captures raw OT data at the edge and works with the NeoFlow No-Code logic engine to visually design ETL workflows, enabling IT/OT data integration without writing code.
Enterprise Context and Semantic Governance
Unified Semantic Model
Build an Ontology of equipment, processes, and enterprise information systems through NeoEdge Matrix, unifying data definitions across IT, OT, and AI
Automated Data Pipelines
A data model-driven mechanism automatically generates data collection and storage workflows and MCP interfaces from a single semantic model, reducing integration costs
Trusted Data Governance
Establish trusted governance through RBAC, HITL, and Audit Log, so enterprise knowledge can be shared across AI applications
Data is more than numbers: it carries relationships and meaning, building a semantic model that the enterprise can trust, that AI can understand, and that grows over time.
Edge AI Deployment and Management
Flexible Model Deployment
Deploy and update AI models flexibly to support model iteration and large-scale deployment needs
Containerization and License Management
Supports containerized models and app management for fast go-live and simple maintenance, with licenses centrally allocated and managed
Multiple Inference Frameworks
Supports NVIDIA TensorRT, Intel OpenVINO, ONNX Runtime, PyTorch, and TensorFlow
Through NeoFlow, AI inference results are combined with multi-source OT data for cross-modal decision-making, reducing false alarms and enabling applications to be quickly replicated to other sites.
Flexible and Open Logic Engine
No-Code Rule Engine
Build local control logic with NeoFlow's drag-and-drop interface for anomaly detection and data preprocessing
Custom Container Deployment
Users can build their own Docker containers to run on NeoEdge X and connect to on-site data
Open SDK
Supports Python and Go (Golang), allowing users to develop Custom Apps
A no-code logic engine designed for the edge, easy to pick up for OT engineers and software developers alike.
AI Agent Closed-Loop Control and Ecosystem Integration
AI Agent Ecosystem Integration
Connect to various AI Agent Hubs through MCP / RESTful API, and integrate existing IT systems such as databases, MES, ERP, and CRM
HITL Human-Machine Collaboration
Combine a Human-in-the-Loop (HITL) Collaboration mechanism so AI Agents complete actions under authorized permissions
Physical Device Control
Use the Command API to feed AI Agent decisions back to physical devices, forming a closed loop from sensing and decision-making to action
Fully on-premises deployment lets AI Agents understand enterprise context and operate physical devices, completing the final step toward Physical AI.
Cybersecurity and Remote Operations
International Security Standards
Complies with the ISO 27001 and IEC 62443 security frameworks, uses TLS 1.3 encryption for communications, and combines TPM 2.0 with mTLS mutual authentication
Access control and auditing
Uses RBAC for role-based permission control, enforces the principle of least privilege, and fully records all operations
Secure Remote Maintenance
Built-in SSH and Proxy features reduce maintenance costs through remote management
AI inference and LLM Agents can both run on-premises, so production data never leaves the factory, balancing real-time performance with security compliance.