What Is Lorel.ai
Last updated: 2026-09-30
This page introduces Lorel.ai: what it can do, key use cases, how it connects to your environment, and its main features.
Overview
Lorel.ai is an operations automation platform that makes infrastructure operations autonomous.
It ingests events and operational data from monitoring tools and cloud services, and manages visualization, detection, incident and task management, AI agent analysis, Runbook execution, notifications, and a record of every response in one place. You can use it to automate incident response, and also to streamline routine operations tasks and job management.
Lorel.ai works alongside your existing monitoring and cloud environments to standardize and automate your operations processes, helping you improve operational quality and reduce operational load.
1. What you can do with Lorel.ai
Automate operations triggered by events
Lorel.ai receives events sent from your monitoring tools and cloud services, and automatically runs follow-up processing based on the rules you configure.
Rules evaluate incoming events and metrics, and incident policies determine how incidents are created. When an incident is created, notifications are sent, and the AI agent or Runbook set in the policy starts automatically as the first response. You don’t need to replace your existing monitoring environment. You can adopt Lorel.ai while keeping the tools and operational workflows you already use.
Automate routine operations and job execution
Define routine operational procedures and jobs as Runbooks, and run them in ways such as the following:
- Run manually
- Run on a schedule from a recurring task
- Start automatically when an incident occurs
- Start a Runbook that an AI agent proposes, once a person approves it
In addition to routine work such as maintenance and periodic checks, you can bring scheduled processes that currently run in an existing job management tool onto the same platform. Redefine them as Runbooks, and run them on a daily, weekly, monthly, or yearly schedule by setting a task to repeat. Managing operational tasks and job execution in a single platform reduces workload and helps keep the quality of your responses consistent.
Analysis and operational support with AI agents
AI agents help you investigate the cause of an incident and understand its impact. Analysis results are recorded on the incident as actions such as adding a note, and the execution history is kept, so your operations don’t depend solely on any one person’s experience.
Visualize operational status (Observability)
In Metrics under Observability, you can list and compare metrics collected from AWS CloudWatch and view them as time series. Checking trends and state changes helps you spot anomalies early and evaluate operational quality. You can also add frequently viewed metrics to your favorites to make routine monitoring more efficient.
Operate cloud environments safely
The cloud action steps in a Runbook normally operate on your cloud environment only within the range of operations you select in Access. You can use Runbooks not only for recovery work but also to automate routine operations and job execution. You can also add an approval step to a Runbook, so that a person approves before execution continues.
An AI agent can read the state of your AWS environment through Access. It cannot run operations that change resources.
Manage operational information in one place
Manage events, metrics, incidents, AI agent analysis results, Runbook execution history, and tasks in one place. Bringing together operational information that tends to be scattered across multiple tools makes it easier to understand the situation and manage operations.
Build up response history and improve
Incident response history, AI agent execution history, and Runbook execution logs are accumulated within their retention periods. Depending on their settings, AI agents refer to related past incidents when they analyze a new one.
By reviewing this history and improving your Runbooks and rules, you can continuously expand the scope of automation and raise operational quality.
2. Key use cases
Lorel.ai brings operational visibility, incident response, and routine operations automation together in a single platform.
Operational visibility
Continuously review operational data collected from your cloud environment to understand the current state of operations.
- Operational data is collected from your cloud environment.
- The collected data is visualized and analyzed.
- You check the status and spot anomalies.
- You make operational decisions and respond.
Automated incident response
When a failure or anomaly occurs, a series of response steps can start automatically.
- Events are received from your monitoring tools and cloud services.
- Rules (detection rules, compound rules, and metric rules) evaluate the events and metrics.
- An incident is created according to the incident policy.
- When a new incident is created, the following run automatically:
- Notification
- AI agent analysis or Runbook execution (one or the other, not both)
- The person in charge investigates and carries out recovery.
- The response is recorded in the incident history.
This standardizes the first response when an incident occurs and reduces missed steps and reliance on individual judgment.
A single incident policy cannot be set to start both an AI agent and a Runbook automatically. Choose one of the two. Also, a detection that has the same policy and the same rule, and the same identifier label values (the labels are set in the incident policy’s “How incidents are split” setting), as an unresolved incident ([Open], [In Progress], or [Escalated]) does not create a new incident. It is added to the existing incident instead.
Automated routine operations
You can also automate routine operational tasks and scheduled processes that currently run in an existing job management tool.
- You start a Runbook manually, or a recurring task starts it on schedule.
- The steps defined in the Runbook (such as cloud operations) run.
- The results are recorded.
- Notifications are sent when a run fails, is waiting for approval, and in similar cases.
Because incident response and routine operations are managed on the same platform, you can unify your entire operations process.
3. System architecture and connection methods
Lorel.ai provides the following two connection methods: one for connecting to external services and one for accessing your cloud environment. You manage both from Integration.
Sources
Sources bring events and operational data in from external services such as monitoring tools and cloud services.
The data you bring in is used for detection by Rules, visualization in Observability, and more. You can use Sources without replacing your existing monitoring environment or operations platform.
Access
Access lets Lorel.ai run operations on your cloud environment safely. You register an IAM role that you created in your own AWS account, and Lorel.ai uses temporary permissions obtained through AssumeRole to perform operations within the range you have allowed.
Runbooks, AI agents, and CloudWatch metrics collection use Access to reach your cloud environment. The cloud action steps in a Runbook normally run only the operations you have selected, and AI agents only read the state of your environment. This automates cloud operations while keeping them safe and governed.
Delivery model
Lorel.ai is delivered as SaaS. You use it from your browser, and you don’t need to install any Lorel.ai-specific software or agents.
Lorel.ai also uses a hierarchy of organizations and projects, so you can keep resources separate for each team or system. The number of projects you can create depends on your plan. For details, see Organizations and Projects Overview and Plans and Pricing.
4. Main features
Lorel.ai offers features that support autonomous, automated infrastructure operations.
| Feature | Description |
|---|---|
| Integration | Manage Sources and Access. |
| Observability | Visualize and analyze metrics collected from your cloud environment. |
| Rules (detection rules, compound rules, and metric rules) | Evaluate events and metrics, and determine detections. |
| Incident Policies | Define how incidents are created after detection, and which AI agent or Runbook starts automatically as the first response. |
| Incidents | Manage the status, response history, and related information of incidents. |
| Tasks | Manage tasks related to incident response and operational work. You can also set tasks to repeat. |
| AI Agents | Run analysis, investigation, and operational support for incidents. |
| Runbooks | Define standard operating procedures and run them automatically or manually. |
| Notifications | Notify you of incident and task updates, Runbooks awaiting approval, and more. |
Related pages
For detailed setup instructions and specifications, see the documentation for each feature.
- Integration (Sources and Access): Integration Overview
- Rules and incident policies: Rules Overview
- Incidents and Tasks: Incidents / Tasks Overview
- AI Agents: AI Agents Overview
- Runbooks: Runbooks Overview
- Organizations and projects: Organizations and Projects Overview
- Roles and permissions: Roles and Permissions
- Security and data handling: Data Handling
- Plans and pricing: Plans and Pricing