Python SDK user bulk creation from CSV throwing 409

General Discussion
1

Stuck on bulk-creating users from a CSV using the Python SDK. I have a loop calling create_user with user_obj, but it fails with 409 Conflict on the second iteration. Is the create_user method synchronous and blocking the next request? I need to handle the rate limiting or use a batch endpoint if available. My current snippet:

for row in csv_reader:
 user_obj = UserBuilder.build_user(row)
 api_response = users_api.create_user(user_obj)

How do I properly sequence these calls without hitting the conflict?

2

TL;DR: 409 is a duplicate email, not rate limiting.

You might want to look at catching the ApiException with status 409 and skipping that row instead of crashing. The create_user endpoint is strict about unique emails, so check if those users already exist in your org before the loop.

3

It depends, but generally… skipping the 409 is a band-aid. You’re hitting the duplicate constraint, not rate limits.

409 Conflict: A user with the same email address already exists.

Check existence first or use the bulk import endpoint instead of looping create_user.

4

If I remember correctly, looping through create_user is a recipe for disaster when migrating from Twilio Studio flows where state is implicit. The suggestion above correctly identifies the 409 as a duplicate email conflict, but you are ignoring the deeper issue: you are not handling the API’s eventual consistency model. When you create a user, the email index update might lag slightly, causing subsequent checks or creates to fail unexpectedly if done too fast. Do not just catch the exception. Use platformClient.users.get_users(email="[email protected]") before creating. If it returns empty, create. If it returns a list, skip or update. Also, add a 100ms sleep between requests. The Python SDK does not auto-throttle. You will hit the global rate limit of 1000 req/min if you spam this loop. Write a wrapper function that checks existence, creates if missing, and sleeps. This mirrors how Twilio Functions handle async DB writes, avoiding the race conditions you are seeing in Architect data actions.

5

How I usually solve this is by pre-filtering the csv with pandas. merge the csv against platformClient.user.searchUsers() results. keep only new emails. this avoids the 409 loop entirely. code: users_df = pd.read_csv(‘data.csv’); existing = pd.DataFrame(api.search()); new_users = users_df[~users_df.email.isin(existing.email)]