Lab 1: Deploying an AIFabric

This lab guides you through deploying an AIFabric resource in the MX-AI ecosystem. An AIFabric enables multiple AI agents to collaborate with each other and use tools from MCP servers.

By the end of this lab, you will be able to:

• Deploy an AIFabric
• Define agents and their roles
• Connect agents to MCP servers and LLMs
• Interact with the system through a UI component

AIFabric Custom Resource

The following YAML defines an AIFabric where agents use OpenAI APIs to generate responses and interact with MCP servers.

myfabric.yaml

apiVersion: odin.trirematics.io/v1
kind: AIFabric
metadata:
  name: myfabric
  namespace: trirematics
spec:
  ui: hub.bubbleran.com/orama/ui/iris
  mcp:
    - name: observability-db
      image: hub.bubbleran.com/orama/mcp/observability-db
      smoAccess: read-only
  llms:
    - name: openai-model
      provider: openai
      model: gpt-4.1-mini
      apiKey: <YOUR-API-KEY>
  topology: supervised
  agents:
    - name: supervisor-agent
      role: supervisor
      image: hub.bubbleran.com/orama/agents/supervisor
      llm: openai-model
      icp: a2a
    - name: smo-agent
      role: worker
      image: hub.bubbleran.com/orama/agents/smo-agent
      llm: openai-model
      smoAccess: read-write
      mcpServers:
        - observability-db
      icp: a2a
  imagePullSecrets:
    - name: bubbleran-hub

Notes

• You may rename the AIFabric, agents, LLMs, and MCP servers by changing the corresponding name field.
• Do not modify the image fields; they are required by the Odin Operator.
• Replace <YOUR-API-KEY> with a valid OpenAI API key. Alternative, use a local model with Ollama replacing the openai-model with the following configuration.

    llms:
      - name: local-model
        provider: ollama
        model: <YOUR-MODEL-NAME>
        baseUrl: <YOUR-OLLAMA-URL>

Deploy the AIFabric

Create the resource with:

brc install aifabric myfabric.yaml

The Odin Operator will deploy all components defined in the AIFabric specification. Wait until all pods in the trirematics namespace reach a Running state. This typically takes a few seconds up to one minute, depending on the number of components and dependencies.

During startup, some pods may be restarted while waiting for required dependencies to become available. This behavior is expected.

Access the UI

The UI is always deployed with name <name-of-aifabric>-ui. In this case, it will be myfabric-ui. In order to use it, follow these three steps:

1. SSH into the BubbleRAN cluster with local port forwarding. This step is not needed if you are working directly on the cluster.

ssh -L 9900:localhost:9900 <your-user>@<your-cluster-ip>

2. Forward the UI port

kubectl port-forward -n trirematics svc/myfabric-ui 9900:9900

3. Open your browser at http://localhost:9900/

Interacting with the Agents

The UI lets you choose which agent to interact with.

In this lab, the AIFabric contains two agents: a Supervisor and an SMO Agent. The Supervisor is designed for more complex scenarios involving multiple worker agents; here, it is included for demonstration purposes.

When you submit a request to the Supervisor, it delegates the task to the SMO Agent, which generates the response step by step. The UI displays the intermediate steps, such as tool usage and text generation, so you can follow the agent’s reasoning process.

Example Questions

After deploying a Network and one or more Terminals, try:

• "What is the TDD configuration of the network?"
• "How many UEs are available?"
• "What is the IMSI of the UE named ue1?"

Since the worker agent relies on a RAG-based approach, try to include keywords present in the Network or Terminal configuration YAML files.

You can also start without any deployment, and use the agents to deploy something:

• "Deploy a new network called testnet with one access network named an.
• "Deploy a UE called user and attach it to the access network an.testnet.

What You Learned

• Deploy an AIFabric resource in the MX-AI ecosystem
• Configure agents, their roles, and supervision topology
• Attach MCP servers to agents and control their access levels
• Configure and use an LLM within an AIFabric
• Use the UI component to interact with agents and observe their execution flow