Extract structured data from documents using AI.

Zendix helps developers extract structured JSON data from invoices, receipts, purchase orders, PDFs, images, forms, and other business documents using asynchronous AI extraction jobs.

How Zendix works

Zendix uses asynchronous extraction jobs. This allows your application to continue running while AI processing happens in the background.

When you create a job, Zendix immediately returns a job ID. You can then either poll the job details endpoint or configure webhooks to receive results automatically.

Your first request

The example below uploads an invoice PDF and creates a new extraction job using a schema ID.

import requests

url = "https://zendix.app/api/v1/jobs/create/"

headers = {
    "Authorization": "Bearer YOUR_API_KEY"
}

files = [
    ("files", open("invoice.pdf", "rb"))
]

payload = {
    "schema_id": 12,
    "document_type": "invoice",
    "client_reference_id": "inv_001"
}

response = requests.post(url, headers=headers, files=files, data=payload)
print(response.json())

const formData = new FormData();

formData.append("schema_id", 12);
formData.append("document_type", "invoice");
formData.append("client_reference_id", "inv_001");
formData.append("files", fileInput.files[0]);

const response = await fetch("https://zendix.app/api/v1/jobs/create/", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_API_KEY"
  },
  body: formData
});

const data = await response.json();
console.log(data);

curl -X POST https://zendix.app/api/v1/jobs/create/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "schema_id=12" \
  -F "document_type=invoice" \
  -F "client_reference_id=inv_001" \
  -F "files=@invoice.pdf"

{
  "success": true,
  "data": {
    "status": "pending",
    "result": null,
    "error": null
  },
  "meta": {
    "job_id": 145,
    "client_reference_id": "inv_001" // present if job was created with a client_reference_id
  }
}

Authentication

All Zendix API requests must include your API key in the Authorization header.

API keys authenticate requests and identify the business making the request.

response = requests.get(url, headers = {
    "Authorization": "Bearer YOUR_API_KEY"
})

const response = await fetch(url, {
  headers: {
    Authorization: "Bearer YOUR_API_KEY"
  }
});

curl https://zendix.app/api/v1/jobs/ \
  -H "Authorization: Bearer YOUR_API_KEY"

Authentication errors

Zendix returns standardized error responses when authentication fails.

{
  "success": false,
  "error": {
    "code": "authentication_failed",
    "message": "Invalid API key."
  },
  "meta": null
}

List jobs

Retrieve a paginated list of jobs belonging to your business.

This endpoint is useful for dashboards, activity history, analytics, monitoring failed jobs, and viewing recent extraction activity.

Query parameters

import requests

response = requests.get(
    "https://zendix.app/api/v1/jobs/?status=successful&page=1&page_size=3&ordering=earliest",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

print(response.json())

const response = await fetch(
  "https://zendix.app/api/v1/jobs/?status=successful&page=1&page_size=3&ordering=earliest",
  { headers: { Authorization: "Bearer YOUR_API_KEY" } }
);

const data = await response.json();
console.log(data);

curl "https://zendix.app/api/v1/jobs/?status=successful&page=1&page_size=3&ordering=earliest" \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "success": true,
  "data": {
    "count": 120,
    "next": "https://zendix.app/api/v1/jobs/?ordering=earliest&page=2&page_size=3&status=successful",
    "previous": null,
    "results": [
      {
        "job_id": 145,
        "status": "successful",
        "client_reference_id": "inv_001",
        "created_at": "2026-05-10T12:45:00Z",
        ...
      },
      {
        "job_id": 146,
        "status": "failed",
        "client_reference_id": null,
        "created_at": "2026-05-10T12:50:00Z",
        ...
      }
    ]
  },
  "meta": null
}

Filtering jobs

Zendix supports filtering jobs by status.

Pagination

Job lists are paginated to improve performance and reduce response sizes.

Create job

Create an extraction job by sending files, text content, or both. Zendix processes the job asynchronously and returns a job ID immediately.

Request fields

Supported combinations

Limits

• Maximum files: 10
• Maximum total upload size: 50MB
• Supported file formats: pdf, png, jpg, jpeg, webp, txt

import requests

url = "https://zendix.app/api/v1/jobs/create/"

headers = {
    "Authorization": "Bearer YOUR_API_KEY"
}

files = [
    ("files", open("invoice.pdf", "rb"))
]

data = {
    "schema_id": 12,
    "document_type": "invoice",
    "client_reference_id": "inv_001"
}

response = requests.post(url, headers=headers, files=files, data=data)
print(response.json())

const formData = new FormData();

formData.append("schema_id", 12);
formData.append("document_type", "invoice");
formData.append("client_reference_id", "inv_001");
formData.append("files", fileInput.files[0]);

const response = await fetch("https://zendix.app/api/v1/jobs/create/", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_API_KEY"
  },
  body: formData
});

const data = await response.json();
console.log(data);

curl -X POST https://zendix.app/api/v1/jobs/create/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "schema_id=12" \
  -F "document_type=invoice" \
  -F "client_reference_id=inv_001" \
  -F "files=@invoice.pdf"

{
  "success": true,
  "data": {
    "status": "pending",
    "result": null,
    "error": null
  },
  "meta": {
    "job_id": 145,
    "client_reference_id": "inv_001" // present if job was created with a client_reference_id
  }
}

Get job details

Retrieve detailed information about a specific job using its job ID. This endpoint is commonly used for polling job status until completion.

Path parameters

import requests

job_id = 145

response = requests.get(
    f"https://zendix.app/api/v1/jobs/{job_id}/detail/",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

print(response.json())

const jobId = 145;

const response = await fetch(
  `https://zendix.app/api/v1/jobs/${jobId}/detail/`,
  {
    headers: { Authorization: "Bearer YOUR_API_KEY" }
  }
);

const data = await response.json();
console.log(data);

curl https://zendix.app/api/v1/jobs/145/detail/ \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "success": true,
  "data": {
    "status": "successful",
    "result": {
      "invoice_number": "INV-1001",
      "total": 250.00
    },
    "error": null,
    "details": {
        "body": "",
        "document_type": "invoice",
        "schema_id": 12,
        "skipped": false,
        "file_count": 1,
        "created_at": "2026-05-10T12:45:00Z",
        ...
    }
  },
  "meta": {
    "job_id": 145,
    "client_reference_id": "inv_001" // present if job was created with a client_reference_id
  }
}

Retry job

Retry a previously failed job. This will reprocess the same job configuration.

Path parameters

import requests

job_id = 145

response = requests.post(
    f"https://zendix.app/api/v1/jobs/{job_id}/retry/",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

print(response.json())

const jobId = 145;

const response = await fetch(
  `https://zendix.app/api/v1/jobs/${jobId}/retry/`,
  {
    method: "POST",
    headers: { Authorization: "Bearer YOUR_API_KEY" }
  }
);

const data = await response.json();
console.log(data);

curl -X POST https://zendix.app/api/v1/jobs/145/retry/ \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "success": true,
  "data": {
    "status": "pending",
    "result": null,
    "error": null
  },
  "meta": {
    "job_id": 145,
    "client_reference_id": "inv_001" // present if job was created with a client_reference_id
  }
}

Delete job

Permanently delete a job and its associated files (if any).

Path parameters

import requests

job_id = 145

response = requests.delete(
    f"https://zendix.app/api/v1/jobs/{job_id}/delete/",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

print(response.json())

const jobId = 145;

const response = await fetch(
  `https://zendix.app/api/v1/jobs/${jobId}/delete/`,
  {
    method: "DELETE",
    headers: { Authorization: "Bearer YOUR_API_KEY" }
  }
);

const data = await response.json();
console.log(data);

curl -X DELETE https://zendix.app/api/v1/jobs/145/delete/ \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "success": true,
  "data": {
    "message": "Job deleted successfully."
  },
  "meta": {
    "job_id": 145
  }
}

Get job by client reference ID

Retrieve a job using its client reference ID. Useful when you do not store internal job IDs.

Path parameters

import requests

job_id = 145

response = requests.get(
    f"https://zendix.app/api/v1/jobs/{job_id}/by-reference/",
    headers={"Authorization": "Bearer YOUR_API_KEY"}
)

print(response.json())

const jobId = 145;

const response = await fetch(
  `https://zendix.app/api/v1/jobs/${jobId}/by-reference/`,
  {
    headers: { Authorization: "Bearer YOUR_API_KEY" }
  }
);

const data = await response.json();
console.log(data);

curl https://zendix.app/api/v1/jobs/145/by-reference/ \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "success": true,
  "data": {
    "job_id": 145,
    "status": "successful"
  },
  "meta": {
    "client_reference_id": "inv_001"
  }
}

Rate limiting

Zendix enforces rate limits to ensure fair usage and system stability.

All API keys belonging to a business can each make a maximum of 100 requests per minute, and collectively make a maximum of 300 requests per minute.

File limits

Zendix enforces strict file upload limits to ensure performance and reliability.

Error codes

All errors follow a standardized structure to make debugging simple and predictable.

{
  "success": false,
  "error": {
    "code": "error_code",
    "message": "Human readable message"
  },
  "meta": null
}

Common error codes

Webhooks

Zendix can send job completion events to your server using webhooks. Webhooks are useful when you want Zendix to notify your backend automatically instead of polling for results.

How to enable webhooks

To enable webhooks, set a webhook URL on the API key in your Zendix dashboard. Zendix uses that URL as the destination for webhook events.

How to prepare your server

Your server must accept POST requests with JSON payloads. It should respond with a successful status code quickly, ideally within a few seconds.

• Use an HTTPS endpoint.
• Accept JSON request bodies.
• Read the raw request body when verifying the signature.
• Return a 2xx response when the webhook is processed successfully.

Webhook signature verification

Zendix signs each webhook using your webhook secret. Your server should verify the signature before trusting the payload.

import hmac
import hashlib

signature = request.headers.get("X-Zendix-Signature")
raw_body = request.body

expected = hmac.new(
    webhook_secret.encode(),
    raw_body,
    hashlib.sha256
).hexdigest()

if not hmac.compare_digest(signature, expected):
    return Response(status=401)

import crypto from "crypto";

const signature = req.header("X-Zendix-Signature");
const rawBody = req.rawBody;

const expected = crypto
  .createHmac("sha256", webhookSecret)
  .update(rawBody)
  .digest("hex");

if (signature !== expected) {
  return res.sendStatus(401);
}

curl -X POST https://your-server.com/webhooks/zendix \
  -H "Content-Type: application/json" \
  -H "X-Zendix-Signature: your_signature_here" \
  -d '{"event":"job.completed"}'

Webhook payload

{
    "event": "job.completed",
    "data": {
        "status": "successful",
        "result": {
            "invoice_number": "INV-1001",
            "total": 250.00
        },
        "error": null,
        "details": {
            "body": "",
            "document_type": "invoice",
            "schema_id": 12,
            "skipped": false,
            "file_count": 1,
            "created_at": "2026-05-10T12:45:00Z",
            ...
        }
    },
    "meta": {
        "job_id": 145,
        "client_reference_id": "inv_001" // present if job was created with a client_reference_id
    }
}

{
    "event": "job.failed",
    "data": {
        "status": "failed",
        "result": null,
        "error": "error_message",
        "details": {
            "body": "",
            "document_type": "invoice",
            "schema_id": 12,
            "skipped": false,
            "file_count": 1,
            "created_at": "2026-05-10T12:45:00Z",
            ...
        }
    },
    "meta": {
        "job_id": 145,
        "client_reference_id": "inv_001" // present if job was created with a client_reference_id
    }
}

Webhook failure tracking

If Zendix fails to send a webhook for a job, the job keeps its extraction result, but the webhook delivery failure is recorded on the job.

• job.webhook_failed indicates whether webhook delivery failed.
• job.webhook_last_error stores the latest webhook error message.

Notes

• Webhooks are sent from the extraction worker.
• Webhook delivery failure does not change the job result.
• Webhook URL is configured at the API key level.