Author: Pavan Mani Deep Y

Passionate for Power Platform. A technology geek who loves sharing the leanings, quick tips and new features on Dynamics 365 & related tools, technologies. An Azure IOT and Quantum Computing enthusiast...

Agentic AI Business Solutions Architect Exam -AB 100 Experience

Hi Folks,

On 14 November, 2025, I took the AB 100 Exam, this post is to share my experience about this exam.

The exam doesn’t look difficult or tricky, it feels like a lot to read in short amount of time. Most of the questions revolved around Copilot Studio, Azure AI Foundry, Azure Services for tracking telemetry, Copilot, Dynamics 365 Customer Engagement, Finance and Operations, Supply Chain Management.

While there is nitty-gritty on using Prebuilt agents and Custom Agents using Azure AI Foundry and Agent Governance, choosing right agent for the need but note that no question came up from AI Builder, Licensing as well.

However there were also scenario based questions on strategy using right tool for the need to build agent e.x. Azure Cloud Adoption Framework, Power Platform Well Architected Framework.

As per Exam NDA, exact exam questions may not be shared publicly, I am sharing my experience so that someone preparing for this exam can use this while preparing for taking this exam.

If you want to learn further, you can go through the below link

AB 100 Collection

Hope it helps..

Cheers,

PMDY

Python + Dataverse Series – #04: Create records in batch using Execute Multiple

Hi Folks,

This is continuation in this Python with Dataverse Series, in this blog post, we will see how can we create multiple records in a single batch using ExecuteMultiple in Python.

Please use the below code for the same…to make any calls using ExecuteMultiple…

import pyodbc
import msal
import requests
import json
import re
import time
# Azure AD details
client_id = '0e1c58b1-3d9a-4618-8889-6c6505288d3c'
client_secret = 'qlU8Q~dmhKFfdL1ph2YsLK9URbhIPn~qWmfr1ceL'
tenant_id = '97ae7e35-2f87-418b-9432-6733950f3d5c'
authority = f'https://login.microsoftonline.com/{tenant_id}'
resource = 'https://ecellorsdev.crm8.dynamics.com'
# SQL endpoint
sql_server = 'ecellorsdev.crm8.dynamics.com'
database = 'ecellorsdev'
# Get token with error handling
try:
print(f"Attempting to authenticate with tenant: {tenant_id}")
print(f"Authority URL: {authority}")
app = msal.ConfidentialClientApplication(client_id, authority=authority, client_credential=client_secret)
print("Acquiring token…")
token_response = app.acquire_token_for_client(scopes=[f'{resource}/.default'])
if 'error' in token_response:
print(f"Token acquisition failed: {token_response['error']}")
print(f"Error description: {token_response.get('error_description', 'No description available')}")
else:
access_token = token_response['access_token']
print("Token acquired successfully and your token is!"+access_token)
print(f"Token length: {len(access_token)} characters")
except ValueError as e:
print(f"Configuration Error: {e}")
except Exception as e:
print(f"Unexpected error: {e}")
#Get 5 contacts from Dataverse using Web API
import requests
import json
try:
#Full CRUD Operations – Create, Read, Update, Delete a contact in Dataverse
print("Making Web API request to perform CRUD operations on contacts…")
# Dataverse Web API endpoint for contacts
web_api_url = f"{resource}/api/data/v9.2/contacts"
# Base headers with authorization token
headers = {
'Authorization': f'Bearer {access_token}',
'OData-MaxVersion': '4.0',
'OData-Version': '4.0',
'Accept': 'application/json',
'Content-Type': 'application/json'
}
# Simple approach: create multiple contacts sequentially
# generate 100 contacts with different last names
contacts_to_create = [
{"firstname": "Ecellors", "lastname": f"Test{str(i).zfill(3)}"}
for i in range(1, 101)
]
create_headers = headers.copy()
create_headers['Prefer'] = 'return=representation'
created_ids = []
print("Creating contacts sequentially…")
for i, body in enumerate(contacts_to_create, start=1):
try:
resp = requests.post(web_api_url, headers=create_headers, json=body, timeout=15)
except requests.exceptions.RequestException as e:
print(f"Request error creating contact #{i}: {e}")
continue
if resp.status_code in (200, 201):
try:
j = resp.json()
cid = j.get('contactid')
except ValueError:
cid = None
if cid:
created_ids.append(cid)
print(f"Created contact #{i} with id: {cid}")
else:
print(f"Created contact #{i} but response body missing id. Response headers: {resp.headers}")
elif resp.status_code == 204:
# try to extract id from headers
entity_url = resp.headers.get('OData-EntityId') or resp.headers.get('Location')
if entity_url:
m = re.search(r"([0-9a-fA-F\-]{36})", entity_url)
if m:
cid = m.group(1)
created_ids.append(cid)
print(f"Created contact #{i} (204) with id: {cid}")
else:
print(f"Created contact #{i} (204) but couldn't parse id from headers: {resp.headers}")
else:
print(f"Created contact #{i} (204) but no entity header present: {resp.headers}")
else:
print(f"Failed to create contact #{i}. Status code: {resp.status_code}, Response: {resp.text}")
# small pause to reduce chance of throttling/rate limits
time.sleep(0.2)
if created_ids:
print("Created contact ids:")
for cid in created_ids:
print(cid)
except Exception as e:
print(f"Unexpected error during Execute Multiple: {e}")
print("Failed to extract Contact ID from headers.")
view raw Python – Dataverse – ExecuteMultipleCreate hosted with ❤ by GitHub

Note: Some part of this code is written with AI, please use with caution, but my main intention is to write about this possibility.

Hope this helps..

Cheers,

PMDY