For Azure Open AI Reinforcement fine tuning, python grader does not work..

Here is the code I ran to test out a very simple python grader according to documentation on how to test these graders

# validate the grader
import os
import json
import requests
from dotenv import load_dotenv

testGrader = {
    "name": "test_grader",
    "image_tag": "2025-05-08",
    "type": "python",
    "source": "def grade(sample, item):\n    return 0.9\n"
}

load_dotenv(override=True)

AZURE_API_KEY = os.getenv("AZURE_API_KEY")
AZURE_API_ENDPOINT = os.getenv("AZURE_API_ENDPOINT")
headers = {"Authorization": f"Bearer {AZURE_API_KEY}"}

payload = {"grader": testGrader}
response = requests.post(
    f"{AZURE_API_ENDPOINT}/openai/v1/fine_tuning/alpha/graders/validate",
    json=payload,
    headers=headers
)
print("validate request_id:", response.headers["x-request-id"])
print("validate response:", response.text)

# run the grader with a test reference and sample
payload = {
  "grader": testGrader,
  "item": {
     "reference_answer": "test"
  },
  "model_sample": "test"
}
response = requests.post(
    f"{AZURE_API_ENDPOINT}/openai/v1/fine_tuning/alpha/graders/run",
    json=payload,
    headers=headers
)
print("run request_id:", response.headers["x-request-id"])
print("run response:", response.text)

The results indicate that while the first request to validate the grader is successful, the second request to run the grader returns an internal error. The output is 0 instead of the expected 0.9, and the response contains "other_error: true" without any further details.

Below is the output received:

validate request_id: "some id"
validate response: {
  "grader": {
    "name": "test_grader",
    "image_tag": "2025-05-08",
    "type": "python",
    "source": "def grade(sample, item):\n    return 0.9\n"
  }
}
run request_id: "some id"
run response: {
  "reward": 0.0,
  "metadata": {
    "name": "test_grader",
    "type": "python",
    "errors": {
      "formula_parse_error": false,
      "sample_parse_error": false,
      "sample_parse_error_details": null,
      "truncated_observation_error": false,
      "unresponsive_reward_error": false,
      "invalid_variable_error": false,
      "invalid_variable_error_details": null,
      "other_error": true,
      "python_grader_server_error": false,
      "python_grader_server_error_type": null,
      "python_grader_runtime_error": false,
      "python_grader_runtime_error_details": null,
      "model_grader_server_error": false,
      "model_grader_refusal_error": false,
      "model_grader_refusal_error_details": null,
      "model_grader_parse_error": false,
      "model_grader_parse_error_details": null,
      "model_grader_server_error_details": null,
      "endpoint_grader_client_error": false,
      "endpoint_grader_client_error_details": null,
      "endpoint_grader_server_error": false,
      "endpoint_grader_server_error_details": null,
      "endpoint_grader_response_validation_error": false
    },
    "execution_time": 0.00041031837463378906,
    "scores": {},
    "token_usage": null,
    "sampled_model_name": null
  },
  "sub_rewards": {},
  "model_grader_token_usage_per_model": {}
}

I would really appreciate if someone could help me figure out the problem here.