Collaboration

Create Workspace

In DigiAI platform, you can create workspaces to isolate data. This ensures that members in different workspaces can only access their own workspace's data. The system is structured into two hierarchical levels: - Organization Level The default Org workspace maintains its isolated data, including AI agents, knowledge bases, and configures the organization-wide shared LLM and Rate Limit (Concurrent Access / Min). - Workspace Level Ensures data isolation between different departments or clients. Scenarios - For Enterprise Use: You can create multiple workspaces to segment different departments of the company. For example, you may create dedicated workspaces for the Marketing team and the Customer Support team, keeping their data isolated. - For Agency Use: Agencies can create separate workspaces for different clients to ensure that each client's data is securely isolated and managed independently. Creating Workspaces To create a workspace, log in as an Owner or Admin . Navigate to the Settings menu and go to the Workspace tab. From there, click on “Create Workspace” , then enter the Workspace Name and a brief Description . After you input this information, the new workspace will be created, and ready for use. When you create a new member in the system, you will have the option to assign specific workspace permissions. This allows you to define which workspace(s) the new member can access and modify. Members will only be able to interact with the workspace(s) you assign to them, ensuring data separation across workspaces. For members that have access to multiple workspaces, they can easily switch between them using the left-hand menu. This allows members to freely navigate between different workspaces to access or edit the data associated with each one, without interfering with the data of other workspaces.

Members Management

Inviting New Members Only Owners or Administrators are authorized to invite new members to the team. To add a new member, the Administrator should click on "Settings" -> "Member" -> "Invite Member". After entering the member's email address and selecting the appropriate permissions for the new member, the invitation process is complete. The invited individual will receive an email to complete their registration. Member Roles and Permissions - Owner: Role Description: The Owner is the initial member who subscribes to DigiAI and establishes the organization by providing the company name. This role carries the highest authority within the organization. Permissions Overview: The Owner has full access to all features and is the only role that can modify subscription plans and manage add-ons, ensuring exclusive control over these administrative functions. - Administrator: Role Description: The Administrator governs the organization or workspace, managing team members and agents. Permissions Overview: Administrators can modify member roles, add or remove both workspaces and members, and manage agents, knowledge bases, tools and workflows. - Participant: Role Description: Regular team members engage primarily as contributors with access to resources within the workspace. Permissions Overview: Members are permitted to utilize agents available within the workspace. They can also edit the existing knowledge bases. Removing Members Administrators hold the exclusive right to remove team members. To carry out this process, administrators should go to "Setting" → "Member", locate the particular member to be removed, and click on "Remove" to finalize the removal. Removing a member from the workspace does not delete their user account from the entire platform. They will still retain access to other workspaces where they hold permissions or may be re-added in the future.

Monitoring

In the Monitoring section, you have the ability to oversee resource usage and assess your agents' effectiveness. The Statistics dashboard provides a robust platform for examining a variety of metrics within the production environment. Key Metrics for Monitoring: LLM Token Usage Keep track of the total number of input and output tokens processed by your Large Language Model (LLM). This includes monitoring the consumption of tokens from your subscribed LLM token add-on. Understanding your token usage can guide necessary adjustments to your subscription to optimize resource allocation. Embedding Token Usage Review the total count of embedding tokens used for Knowledge Base. Monitor how these tokens consume your subscribed embedding token add-on. This helps ensure you have the necessary resources for efficient operations and informs decisions regarding potential add-on adjustments. Peak Concurrent Access Observe peak concurrent access per minute to assess the maximum number of users interacting with your agent at once. This metric is crucial for determining if your subscription's rate limit add-on needs adjustment, ensuring consistent performance even during high-demand periods. Total Interactions This metric captures the average number of continuous interactions during a single user session. For instance, if a user engages in a 10-round Q&A session with the AI, it is recorded as 10 interactions. This data serves as an indicator of user engagement levels. Total Conversations Examine the total count of AI conversation sessions. Each new session counts as one, regardless of the number of message exchanges within it. Conversations related to debugging are not included. Active Users This figure represents the count of unique users who have engaged in meaningful interactions with the AI, denoted by having multiple question-and-answer interactions.

Alert Setting

To set up alert notifications for your AI Agent's resource usage, follow these steps. Owners or Admins can create and manage alerts to ensure smooth operations: Access the Alert Settings Go to the Monitoring page and navigate to the Alert Setting tab. Click on New Rule to create a new alert rule. Configure the Alert Rule - Select the Model : Choose the LLM or Embedding model for which you want to set up an alert. - Set the Token Threshold : Specify the token threshold that, when reached or exceeded, will trigger the alert. This threshold will be set for input, output or embedding tokens. - Choose Email Recipients : Decide which Admins should receive an email notification when the alert is triggered. - Click Save to activate the alert rule. Alert Activation Once set, if the input, output or embedding tokens for the selected model fall below the specified threshold, an email notification will be sent to the identified Admins. Notification Frequency Each alert rule is designed to notify the specified Admins only once within a 7-day period, even if the token usage remains below the threshold continuously.