Skip to main content

AI-powered App Review

Overview

Gem’s AI-powered App Review is an add-on within Gem’s standard Application Review product that helps recruiters find top applicants faster using AI inbound ranking. When enabled, Gem AI can consistently rank thousands of applications based on the criteria you input, then surface the best matches to the top of your stack.

AI inbound ranking can be turned on optionally for individual job requisitions in Gem’s Application Review product, and we’ll use AI to score applicants' resumes based on your input criteria, producing an overall score. We then use the score to prioritize the order in which the applicants are shown. We do not make any automated decisions on advancing or rejecting those applicants in the product.

 

Eligibility

  • Your team has access to Gem CRM and Application Review.
  • Your team uses one of the following ATSes: Greenhouse, Workday, SuccessFactors, and iCIMS.

Please reach out to your dedicated Gem Customer Success Manager to learn more on how to get started!

 

Setting up AI ranking for a job

  1. When viewing a job in Application Review,  click Enable AI ranking (this step does not trigger any product behavior).
    • We made this configurable on a per-job basis to give you more control.

 

Tip! Only the jobs with the ✨ in the list of jobs have AI ranking enabled.



  1. For each job, a user inputs “scoring criteria” that the AI will use to score resumes. You can ‘import scoring criteria’ if you have a related AI sourcing search. 

 

How scoring works

We strip the applicant-submitted resume (and only the resume) of the candidate’s full name, email address, and LinkedIn profile. Then, we ask the AI to score the provided resume against each of the criteria you have written, which produces a score for how well the resume fits the criteria. Finally, there’s a weighting mechanism (provided by the ⭐: starred = 5x the weight) that we use to produce a weighted average % score that’s displayed (ex. 90%).

 

Things to note with AI ranking

We’ll warn you (example below in yellow) if the criteria 1) is not written properly (i.e. not a full sentence), or 2) criteria includes highly subjective soft skills. We continue to add to these proactive guardrails to reduce introducing bias to the process or to the AI scoring process.

 

Please keep in mind that AI is not perfect, and this tool is meant to aid you in their inbound review - not replace it. AI can make mistakes in producing scores and we’re open to feedback for when it doesn’t get it quite right.

 

Reviewing AI scores

There are two places in Gem where you can see the AI scores:

  • In the list to the left (now can be sorted by ‘AI ranking’ % score).
  • In each resume - each resume will get an ‘AI review’ panel that displays the score produced by AI ranking.

 

Best practices for writing scoring criteria

  • Use AI ranking with pre-existing App Review filters (such as ‘keywords’) when needed. This is especially helpful for prioritizing applicants that have specific skills relevant to a certain role (keyword search) and meet your inputted qualifications (AI ranking).
    • For example: “Keyword filter” > Filter out only folks with ‘revenue’ listed on their resume (since this is a non-negotiable for the role)
  • Use AI ranking as a way to specify criteria that is hard to express in a Boolean filter, or so that you don’t need to type everything out. This helps to further sort the list.
    • For example: imagine the below screenshot with 100s in volume
      • Experience at a top tech company
      • Familiarity in Renewals, Expansion, Customer Lifecycle, Retention, Account Planning, Strategic Selling

 

Have any issues or questions on this topic? Please feel free to contact your dedicated Gem Customer Success Manager directly or our Support team at support@gem.com.

Was this article helpful?