Skip to main content
Beta access required. Beam is currently in beta. Contact support@allium.so to get access before starting.

​
Create your first pipeline

1

Open the pipeline builder

Navigate to app.allium.so/build/beam and click New to create a blank pipeline, or Start with Template for a pre-built example.
2

Configure your source

Select a blockchain and entity type to stream. For example, Polygon + log to stream all Polygon logs.
Beam source configuration
3

Add a filter

Add a Set Filter to narrow down to specific events. Enter the contract addresses you want to monitor and select the field to match against.For example, filter to USDC on Polygon by entering the USDC contract address and matching on the address field:
Beam set filter configuration
4

Add a sink

Add a Kafka sink and give it a name (e.g., usdc-logs). This determines where your filtered data gets delivered.
Beam Kafka sink configuration
5

Deploy

Click Deploy. Beam creates Kafka topics, provisions credentials, and spins up workers — typically in under 30 seconds.
6

Consume your data

After deployment, click Get Cooking to view your Kafka credentials and ready-to-use code snippets.
Beam deployment credentials
from confluent_kafka import Consumer

conf = {
    "bootstrap.servers": "<bootstrap-server>",
    "group.id": "<consumer-group>",
    "security.protocol": "SASL_SSL",
    "sasl.mechanisms": "PLAIN",
    "sasl.username": "<username>",
    "sasl.password": "<password>",
    "auto.offset.reset": "latest",
}

consumer = Consumer(conf)
consumer.subscribe(["beam.<config_id>.usdc-logs"])

while True:
    msg = consumer.poll(1.0)
    if msg is None:
        continue
    print(msg.value().decode("utf-8"))
Replace the placeholder values with the credentials shown in the Get Cooking panel.

​
Verify your deployment

After deploying, check worker health in the Beam UI. You should see:
  • Total workers — number of worker pods running
  • Healthy workers — workers processing data normally
  • Unhealthy / crashing / OOM — should all be 0
Use the Metrics panel to confirm messages are flowing (1h, 1d, or 1w time range).

​
Updating a pipeline

To change your pipeline config, edit the source, transforms, or sinks in the UI and click Deploy again. Deployment is idempotent with zero downtime — changes apply in place.
Do not teardown before redeploying. That causes unnecessary downtime. Just redeploy directly.

​
Next steps

Configuration reference

Full reference for sources, transforms, and sinks

Examples

Copy-paste pipeline configs for common use cases