Hey everyone — has anyone dealt with IXP?
I’ve been exploring a few UiPath capabilities that are new to me recently, specifically IXP (Intelligent Xtraction and Processing). I’m discovering it on extraction information from Resumes as part of my learning and research into AI-driven document processing.
Following the official academy course and available technical guides, the expectation is that after validating predictions and training the model, the Project score should update accordingly. Instead, after submitting annotations, I’m getting a hard failure: “Failed to predict extractions”,
and the right-side panel now shows previously confirmed fields (Applicant’s First Name, Last Name, etc.) as “Marked as missing” — even though the values are clearly visible and were already submitted/confirmed.
Digging into DevTools, the actual API response behind this is:
404
{
"status": "error",
"message": "The requested user model doesn't exist."
}
It looks like the backend can’t find or resolve the model tied to this project anymore (possibly a desync after taxonomy edits or model config changes).
My questions for the team:
Has anyone encountered this error “The requested user model doesn’t exist” on IXP (searching and talking to Claude/Gimini didn’t yield any answers) or am I missing something specific in the project/account settings during configuration?
I have a suspicion this might be tied to an account provisioning limitation. My current UiPath account was registered many years ago with a personal Gmail address (other services/products are still working normally), which is already blocking me from opening an official Customer Portal support ticket (“Registration using public email domains is not permitted…”).
Mostly looks lime you are using structured extraction from the discriptuon you have
And marked as missing comes when enough samples are not trained
Also for the document you are looking for..looks like resume its better to go with unstructured documents which dors mot need annotationa nd would work on prompts..as there are no identifiers and fixed formats that would make sense
But if you want to learn go with standard invoices for structured docs
For resume screening, I would prefer using Generative AI or Agentic AI approaches rather than training an IXP model, especially if the resume formats vary significantly.
IXP can work well when documents have a relatively consistent structure and you have enough samples for training. However, resumes often come in many different layouts, making them a challenging use case for a traditional extraction model.
Regarding your issue, since you’re using an individual UiPath account, you can raise a support request through the Learning Support team using the link below:
Here is recommended to annotate at least 10 examples for each field so that the model can provide you with meaningful validation statistics, but it is impossible due to quotas.
If you need to learn these advanced things, then better start a enterprise trail. but it has only 60 days after which it goes to free, to get community again you might need to delete the org completely and re create one in community, else use a separate email and create community
