Skip to main content
All CollectionsAI Modules
How to test HrFlow.ai Resume Parsing?
How to test HrFlow.ai Resume Parsing?
seiv mouhidine avatar
Written by seiv mouhidine
Updated over a year ago

Using AI-powered prediction models can be very helpful and save you a lot of time and effort, but before you put this new mechanism in place, it is important to test and verify the quality of the AI algorithms you will implement.

At HrFlow.ai, we are aware of this challenge, and that's why we put in place a process that allows the customer to audit the results of all our algorithms before deploying them.

This process allows the customer to test our AI algorithms on their own data and evaluate their performance to ensure the quality of the results.

For HrFlow.ai's Resume Parsing technology, the approach you take hinges on three factors:

  • The technical proficiency of the tester.

  • The depth of insight anticipated from the test.

  • The comprehensiveness expected from the test.

  1. Fast and non-technical test

    You can try HrFlow.ai Resume Parsing without signing up or possessing technical expertise. Simply follow these steps:

    Your results will be delivered to your email.
    Please note that this method allows testing of only one resume at a time.

  2. Self-serve & technical test

    You can try the HrFlow.ai Resume Parsing API by signing up. Simply follow these steps:

    HrFlow.ai Workspace

    Please note that this method requires a credit card, but you can parse up to 100 resumes/month for free.

  3. Exhaustive & large batch test

    Our technical support team can work with you to generate extensive statistics about the performance of our Resume Parsing for large datasets of Resumes. Our limit is set to 1,000 resumes/customer.

    Simply follow these steps:

    • Use Wetransfer to forward your sample dataset to support+parsing@hrflow.ai

    • In less than 24 hours, our team will provide you with an in-depth Excel spreadsheet detailing the Resume Parsing performance.

    Resume parsing performance metrics.

    In this spreadsheet:

    • Each row corresponds to a parsed profile.

    • Each column denotes a specific performance metric.

    Performance metrics include:

    • Binary metrics include name, email, start date, end date, and more.

    • Aggregated metrics like info_score, exp_score, edu_score

    • Count metrics, for instance, skills.

    Please be aware that our actual performance metrics might be higher. In fact, they represent a conservative estimate, as the data we're evaluating the model against might not be present in the resume.

Did this answer your question?