Skip to main content
OpenTrain AI screens candidates automatically before they can submit a proposal. When a candidate starts a proposal for your job, they are routed through an AI-driven interview that captures structured answers based on the requirements you configure in the job wizard.
The AI interview is required for all jobs that have recruiting screening fields configured. Candidates cannot submit a proposal until the interview is complete and all required fields have been captured.

​
Configure AI interview requirements

You set screening requirements in Step 5 of the job post wizard.

​
AI interview prompt

On Step 5, Page 1, you write the instructions the AI interviewer follows with every applicant. The platform generates a suggested prompt from your job description — you can accept it, edit it manually, or regenerate it at any time. The prompt should describe:
  • Qualifications and credentials to verify
  • Specific experience the interviewer should probe for
  • Any role-specific questions (availability windows, tools, language skills)

​
Screening fields

On Step 5, Page 3 (available on recruiting-configured jobs), you define individual fields the interviewer will ask candidates about. Each field has:
PropertyDescription
LabelThe display name shown in the screening table and CSV export
DescriptionWhat the interviewer should ask or verify
RequiredWhether a missing answer blocks interview completion
Response typeShort text, long text, URL, number, or single select
Common field types:
  • LinkedIn URL — The interviewer accepts linkedin.com/in/... URLs.
  • Years of experience — The interviewer cross-checks against the candidate’s resume and flags mismatches.
  • Availability — A short-text or single-select field describing the candidate’s weekly availability or timezone.
  • AI training / data labeling experience — Captures direct AI training, annotation, RLHF, or evaluation experience.

​
How candidates experience the AI interview

When a candidate opens a proposal for your job:
  1. They are routed to the AI interview before reaching the proposal form
  2. The AI interviewer asks questions based on your configured requirements, in a conversational format
  3. Answers are collected as the candidate responds
  4. When all required fields are collected, the interview completes and the candidate can submit their proposal
The interviewer uses resume and profile data to cross-check answers. Mismatches are flagged in the screening table but do not block the candidate from completing the interview.

​
Review screening answers

After candidates complete interviews, you can review their structured answers in the Folder Screening tab.
1

Open the folder

From My Jobs, click the folder name to enter the folder view.
2

Go to the Screening tab

Select the Screening tab inside the folder. The screening table shows one row per candidate, with columns for each structured field you configured.
3

Filter and search

Use the filter buttons to show all candidates, shortlisted candidates, candidates with complete interviews, or those with missing required fields. Use the search box to find a specific candidate by name or answer value.
4

Manage columns

Click Manage columns to choose which fields appear in the table. The column selection is saved per folder.
Resume review signals — columns like resume match score and resume review status are surfaced in the table when the interviewer detected a mismatch between a candidate’s answers and their uploaded resume. These are informational and do not block proposal submission.

​
Export screening data

To share candidate screening data with a client or external stakeholder:
  1. Open the folder and go to the Screening tab
  2. Apply any filter you want reflected in the export (e.g., Shortlisted or Complete)
  3. Click Export CSV
The downloaded file contains one row per candidate and one column per screening field, using the export labels you defined when configuring each field. URL fields and AI training experience fields use their normalized display values in the export.