Stream cohort | Sarvam API Docs

import requests
url = "https://apps.sarvam.ai/api/scheduling/v1/orgs/org_id/workspaces/workspace_id/deployments/deployment_id/cohorts/stream"
payload = {
    "name": "known-callers-batch-1",
    "users": [
        {
            "user_phone_number": "+919876543210",
            "user_identifier": "CUST-001",
            "app_variables": { "customer_name": "Rahul" }
        }
    ]
}
headers = {"Content-Type": "application/json"}
response = requests.post(url, json=payload, headers=headers)
print(response.json())

{
  "name": "known-callers-batch-1",
  "cohort_id": "known-callers-batch-1-xxxxx",
  "status": "processing",
  "source_type": "file_upload",
  "created_by": "user@example.com",
  "created_at": "2026-08-05T10:00:00Z",
  "updated_at": "2026-08-05T10:00:00Z",
  "result": null,
  "updated_by": null,
  "cohort_category": "cohort"
}

Stream a cohort as JSON with two fields:

name - Cohort name (1-50 characters)
users - Array of user records (1-1000)

When to use

Stream cohort adds known callers to an inbound deployment programmatically. When a matched phone number calls in, the agent receives personalized variables for that contact.

CRM integration - keep caller context in sync from Salesforce, HubSpot, Zoho, or your own database
Webhooks and automation - update caller profiles when account data changes in your systems
Incremental batches - add or refresh contacts over time without CSV uploads

For CSV files, use Upload cohort instead.

Processing is asynchronous - poll Get cohort until status is completed or failed.

Stream a cohort as JSON with two fields: - `name` - Cohort name (1-50 characters) - `users` - Array of user records (1-1000) **When to use** Stream cohort adds known callers to an inbound deployment programmatically. When a matched phone number calls in, the agent receives personalized variables for that contact. - **CRM integration** - keep caller context in sync from Salesforce, HubSpot, Zoho, or your own database - **Webhooks and automation** - update caller profiles when account data changes in your systems - **Incremental batches** - add or refresh contacts over time without CSV uploads For CSV files, use **Upload cohort** instead. Processing is asynchronous - poll **Get cohort** until `status` is `completed` or `failed`.

Path parameters

org_idstringRequired

workspace_idstringRequired

deployment_idstringRequired

Request

This endpoint expects an object.

namestringRequiredformat: "^[\w\- ]{1,50}$"<=50 characters

Cohort name (1-50 characters). Letters, numbers, spaces, underscores, and hyphens only.

userslist of objectsRequired

List of user records (1-1000).

Response

Successful Response

namestring

Name of the resource

cohort_idstring

Unique identifier for the cohort

statusenum

Current status

source_typeenum

How the cohort was uploaded

created_bystring

User who created this resource

created_atdatetime

Timestamp when the resource was created (ISO 8601)

updated_atdatetime

Timestamp when the resource was last updated (ISO 8601)

resultobject or nullOptional

Processing result with record counts

updated_bystring or nullOptional

User who last updated this resource

Errors

422

Unprocessable Entity Error

Stream a cohort as JSON with two fields:

name - Cohort name (1-50 characters)
users - Array of user records (1-1000)

When to use

Stream cohort adds known callers to an inbound deployment programmatically. When a matched phone number calls in, the agent receives personalized variables for that contact.

CRM integration - keep caller context in sync from Salesforce, HubSpot, Zoho, or your own database
Webhooks and automation - update caller profiles when account data changes in your systems
Incremental batches - add or refresh contacts over time without CSV uploads

For CSV files, use Upload cohort instead.

Processing is asynchronous - poll Get cohort until status is completed or failed.