Skip to content

SenseOn AI Architecture

SenseOn built its AI around Horus. Horus is an AI orchestrator and chat interface. It coordinates a set of specialist agents on top of the built-in detection of the platform. You use chat to communicate with Horus. In the background, Horus sends your request to the correct agent and gets the data that it needs. Then Horus shows the results as cases and insights that you can act on.

This page tells you how these parts work together.


The architecture, in four layers

The SenseOn AI uses an architecture with layers. Data moves up through the layers:

  • The platform ingests and unifies the telemetry.
  • The platform detects and correlates the telemetry into cases.
  • Horus coordinates specialist agents. These agents work with that data for you.

1. Data Fabric: the data layer

The Data Fabric ingests all telemetry across endpoint, network, cloud, identity, and third-party tools. It unifies this telemetry into one foundation that you can query. Each layer above the Data Fabric works from this unified view of your environment.

2. Detection & correlation: built-in

This layer is part of the platform. It continuously correlates telemetry and automatically creates cases. It is a core behaviour of the platform, not a named or marketed capability. Custom agents can extend this layer.

3. Horus: AI orchestration & chat

Horus is the AI orchestrator and chat interface of SenseOn. It coordinates the specialist agents, and it is on top of the built-in detection of SenseOn.

Horus is between the analysts and the agents that do the work. You ask a question or set a goal in chat. Horus sends it to the correct agent. The results then show as cases, and analysts act on these cases.

4. Specialist agents

SenseOn designed each specialist agent for one part of the security workflow. Horus coordinates each agent. Some agents also show as dedicated modules in the platform. You can get access to all agents through Horus chat.

Agent What it does Where you find it
Resolve Investigates and closes cases autonomously. Gives verdicts that evidence supports. Investigate and Experience (Resolve)
Hunter Hunts telemetry and builds queries. Writes, debugs, and explains CHSQL. Hunt Lab, Horus chat (Hunt Lab)

For details about how SenseOn makes AI Case Summaries (narrative summaries for each case), refer to AI Case Summaries. That page also tells you how the generative AI of SenseOn handles data and privacy.


How it fits together

  • You work through Horus. Horus is the one chat interface for all of the SenseOn AI. If you ask it to investigate a case or hunt across telemetry, it coordinates the correct specialist agent.
  • Agents work from the same data. Each agent uses the unified Data Fabric and the built-in detection and correlation of the platform. Thus, the results of all agents are consistent, and they come from the same telemetry.
  • Results show as cases. The platform automatically changes correlated activity into cases. Horus and its agents enrich, investigate, and explain those cases. Your analysts make the final decision.