AI Calibration
AI Calibration helps recruiters create strict knockout filters by improving how must-have qualifications are written inside the Job Description. Instead of treating every requirement as a general preference, it turns the Job Description into a structured screening filter that the AI can act on precisely.
This improves candidate quality, reduces manual rejection work, and ensures recruiters spend time only on profiles that genuinely match the role requirements.
AI Calibration works only based on the Job Description and the skill evidence available in the candidate's resume and public application data. If a required qualification is not clearly written in the JD or cannot be found in the candidate's visible information, the AI cannot consider it during screening. Candidates may be marked as Disqualified based on knockout rules even when the missing information is simply not visible in their profile.
When to Use AI Calibration
AI Calibration is most effective when too many low-fit candidates are passing through initial screening, or when important role requirements are not being treated as strict deal breakers. It works best for roles where specific experience, tools, certifications, domain knowledge, or location requirements need to be clearly enforced during shortlisting.
How to Set Up AI Calibration
In the Skima AI App
Navigate to 'Jobs' from the left sidebar and open the job where you want to strengthen candidate screening.
Click the 'AI Calibration' button in the job page header. This opens the calibration panel on the right where knockout conditions are created and managed for that specific role.
If no knockout rules have been added yet, the panel will show an empty state. This means the system is waiting for you to define must-have qualifications in the Job Description.
Navigate to the 'View JD' tab and click 'Edit Job Description'.
Add must-have requirements using clear, specific, and measurable language. Use direct wording like "must have", "mandatory", "required", or "strictly" so the AI can identify them as knockout conditions during screening.
The Job Description you edit here is visible to candidates on your public job posting. If you want to add internal-only screening conditions that candidates cannot see, scroll down to the 'Additional Job Description' section. Anything added there is used exclusively for internal AI screening. Refer to the Create Job guide for more detail on how the JD and Additional JD sections work together.
Click 'Save and Regenerate'. Skima AI will reassess the role and update the knockout filters in the calibration panel based on the updated JD.
In the AI Calibration panel, review the configured knockout conditions and select the ones that should disqualify candidates who do not meet them. This separates strict deal breakers from preferred qualifications.
Click 'Apply and Score Candidates'. This applies the calibration logic and refreshes candidate scoring for the job.
Return to the 'View Candidates' tab. Candidates who fail the selected knockout conditions will show a Disqualified tag with a detailed explanation of which condition failed and why they were removed from the qualified pool.
In the Chrome Extension
AI Calibration is configured in the Skima AI app only. The Chrome Extension reflects the calibrated screening results but does not allow setup or editing.
- Install and log in to the Skima AI Chrome Extension.
- Open any job where AI Calibration has already been applied.
- Candidates matching your knockout criteria appear at the top of the list. Disqualified candidates are pushed below them automatically.
- Each disqualified candidate shows a Disqualified tag with an explanation of which knockout condition they failed.
JD Writing Guide
The quality of your knockout filters depends entirely on how clearly the must-have requirements are written in the JD. Below are examples for the most common screening scenarios.
Minimum Years of Experience
Write the number directly so the AI can apply it as a strict threshold.
Must have a minimum 3 years of B2B SaaS sales experience.
Candidates are required to have at least 5 years of experience in full-cycle recruitment.
Strictly 4+ years of hands-on experience in enterprise account management.Specific Tool or System Knowledge
Name the exact platform instead of using broad wording like "comfortable with tools."
Must have hands-on experience using Workday for recruitment operations.
Candidates are required to have experience working with Salesforce CRM in a sales role.
Experience with Greenhouse is mandatory for this role.Industry Experience
State the industry directly rather than writing a general preference.
Must have prior experience working in the healthcare SaaS industry.
Candidates are required to have experience selling to enterprise clients in the logistics domain.
Candidate should strictly have experience in B2B fintech sales.Certifications or Licenses
Write the certification as a strict condition when the role cannot move forward without it.
Candidates must mandatorily hold a valid Six Sigma Green Belt certification.
Project Management Professional certification is mandatory.
Must have a current AWS Solutions Architect certification.Location, Shift, or Availability
Write location and availability conditions clearly when they are non-negotiable.
Must be available to work in the Mumbai office five days a week.
Candidates must be willing to work night shifts in the US time zone.
Notice period should be strictly 30 days or less.Best Practices
Separate must-haves from nice-to-haves. When both are mixed in the JD, the AI may treat a soft preference as a strict condition or miss an important requirement entirely. Write mandatory conditions in the JD Summary and keep preferences in a separate line or section.
Use specific and measurable language. Broad wording like "good experience" or "strong background" does not give the AI enough detail to apply an accurate knockout filter. Mention exact years, tools, industries, or certification names wherever possible.
Keep knockout rules realistic. Only add conditions that would truly remove a candidate from consideration. If too many requirements are marked as strict, the shortlist may shrink too far and strong candidates may be filtered out too early.
Update calibration when the role changes. The AI follows the written JD. If the hiring manager updates requirements, the JD and calibration should be refreshed immediately to keep screening accurate.
If you are unsure how to write knockout criteria for your specific domain or role type, reach out to [email protected] for guidance.