When 10,000 resumes meet 100 JDs, find the candidates worth meeting.
Last updated Reviewed by Hab Business Solutions
MinMaxHR is the hiring decision system. CandidRanker is the proprietary ranking engine inside it: it scores every applicant against the role across eight dimensions with the evidence attached, so your team can find the best fit and build a stronger team without replacing the ATS.
CandidRanker is the proprietary 8-dimensional scoring engine that powers the MinMaxHR workflow.
- 8-dimensional candidate scoring
- No candidate is ever auto-rejected
- Every decision logged with an owner and a reason
- Data stored and processed in India (asia-south1)
- One bounded pilot, measured against your baseline
MinMaxHR · CandidRanker engine
Evidence-based hiring decisions at scale
Sr. Product Designer
100 JDs · 10,000 resumes · one evidence standard
Best-fit candidate
Evidence-backed reasoning
Find the best fit for the role, then build a winning team with the evidence visible to the people making the decision.
Ranked candidates
8 dimensions
- 94Shortlist
Aria Chen
Product Designer
- 91Review
R. Subramanian
Product Designer
- 87Review
Priya Iyer
Product Designer
Evidence
What matched · what is missing · why
Human decision
Recruiter decides · system records
Short answer
What is MinMaxHR?
What it does
Capabilities that map to business outcomes.
01
Hiring KPI measurement
Time-to-decision, time-to-fill, and funnel conversion baselined and tracked, so improvement is provable rather than anecdotal.
02
Ranked shortlists with evidence
CandidRanker scores applicants across eight dimensions and shows what matched, what is missing, and why.
03
Decision governance
Four explicit decision states, a named owner and written reason on each, and a scoring audit trail covering every weight and filter change.
04
Works with your ATS
No rip and replace. MinMaxHR sits alongside your existing stack and upgrades the decisions, not the plumbing.
05
Consistent evaluation criteria
The same criteria applied identically across roles and recruiters, so hiring quality stops depending on who happened to open the resume.
06
Leadership visibility
One governed view of hiring performance for founders, HR heads, and talent leaders, with minimum effort from the team producing it.
The decision loop
CandidRanker, the proprietary engine that powers the workflow.
From raw applications to a defensible shortlist, the system handles repeatable evaluation while a person keeps the decision.
Human decision layer
The system ranks and explains evidence. Your hiring team decides who moves forward.
- 01
Upload
Add resumes and job descriptions in PDF, Word, text, HTML, or image and scanned formats.
- 02
Identify and check
Identify the document type and check whether the content is usable for evaluation.
- 03
Define
Set mandatory and optional requirements, terminology, and scoring preferences for the job.
- 04
Rank
CandidRanker evaluates candidates against the specific job description across eight evaluation dimensions.
- 05
Inspect
Review matches, missing requirements, experience, education, certifications, projects, and similarity signals.
- 06
Decide
Shortlist, select or reject with a reason. The current decision can also be cleared.
- 07
Interview
Generate interview preparation for shortlisted or selected candidates, tied to the job and the evidence.
- 08
Report
Share decision information through the available report and export formats.
Do not ask a buyer to trust a number. Show what matched, what was missing, which dimensions contributed, and what evidence supports the result.
Before the ranking
Input quality comes before match quality.
A strong ranking needs usable evidence. Job description quality and resume quality are assessed separately from how well a candidate fits.
Usable before matched
Documents are checked for whether they are complete enough, readable, extractable, correctly identified, and coherent before any candidate fit is calculated.
Security by default
Uploaded documents are malware scanned by default with ClamAV. A workspace can switch scanning off, and files are then treated as unscanned.
Quality routing
Poor, incomplete, unreadable or ambiguous inputs are surfaced for human review rather than producing false confidence.
Control the decision logic
Make the ranking useful, not just fast.
The value is not the score alone. The value is knowing what the score means and what your team should inspect next.
Boolean skill groups and aliases
Group equivalent terms so different wording does not create avoidable misses.
Mandatory and optional
Separate must-have requirements from useful extras. Missing mandatory requirements stay visible.
Workspace terminology
Use your organisation's own language consistently across future rankings.
Degree-aware matching
Recognise supported degree variants and relevant education context instead of relying only on exact text.
Skill depth and recency
Look beyond whether a skill appears. Inspect evidence of duration and recency in work history.
Employment gaps
Surface longer gaps as neutral facts for discussion rather than treating them as an automatic penalty.
After the shortlist
From shortlist to interview, on the same evidence.
Once a candidate is shortlisted or selected, the same job description and candidate evidence turn into interview preparation. Reports export as PDF, Excel, Word, HTML.
Verify
Questions can verify claimed strengths against employers, projects, dates, and other evidence.
Probe
Questions can focus on missing requirements, with mandatory gaps made clear.
Discuss
Career gaps become neutral discussion points instead of automatic penalties.
Score
An interview rubric can align the conversation with the same job requirements used in ranking.
Who it covers
Blue collar, grey collar, and white collar.
The same job title means different things across workforces, so the criteria change with the work rather than the wording.
Blue collar
Yes. Blue collar hiring is usually trade, plant and field work where the resume is short, the certification matters more than the prose, and the same role is filled repeatedly at volume. MinMaxHR is configured to weight licences, tickets, machine and equipment familiarity, shift history and physical-site experience, and to read the scanned or photographed documents these candidates actually submit.
Grey collar
Yes. Grey collar roles combine a technical trade with a qualification or a supervisory scope: technicians, maintenance and quality engineers, healthcare support, logistics supervisors, field service. These are the roles a keyword search fails most often, because the title is generic and the actual capability sits in the project detail. MinMaxHR scores the skill depth and the evidence under it, not the job title.
White collar
Yes, and this is where the title problem is worst. A specialist and a generalist can carry the same job title and the same platform name while doing entirely different work. MinMaxHR is configured with the distinctions that matter for the role, so the evidence for hands-on delivery is scored separately from adjacent exposure to the same domain.
Industry criteria
Configured per role, not shipped as a fixed catalogue.
Yes. Criteria are configured per workspace and per role rather than shipped as a fixed catalogue. The same job title means different things in manufacturing, IT services, healthcare and retail, so the criteria set, the terminology catalogue and the weighting are defined for the role you are actually filling.
Manufacturing and plant hiring
Module and process experience, plant and production exposure, maintenance and quality scope, trade tickets. The distinction between implementing a system and operating it is written into the criteria.
IT services and technical hiring
Hands-on build evidence separated from functional, support and coordination work carrying the same platform name. Depth and recency of a named stack scored against the requirement.
Healthcare and clinical support
Registrations, licences and clinical scope treated as mandatory criteria where the role requires them, with employment history reviewed by a person rather than filtered by a rule.
Retail, logistics and high-volume intake
Repeatable criteria applied identically across a large seasonal pool, so the hundredth applicant is evaluated on the same standard as the first.
Honest fit
Who it's for, and who it isn't.
Trust is built by what a vendor admits. Here's both sides, plainly.
A strong fit if
- HR and talent teams facing applicant volume they cannot read honestly
- Staffing and recruitment agencies that win or lose on shortlist speed and quality
- Enterprises that need consistent, auditable hiring decisions across role families
- Founders hiring their first 10 to 50 people without a dedicated talent team
Not the right tool if
- You need job posting and applicant tracking only. A classic ATS already does that.
- You want humans removed from hiring decisions. We deliberately do not build that.
- You are looking for an agency to source candidates for you.
Questions
Asked and answered.
Is MinMaxHR an ATS?
No. MinMaxHR is a decision intelligence layer for recruitment. It sits next to your applicant tracking system rather than replacing it: the ATS keeps tracking applicants, MinMaxHR improves the decisions made inside that pipeline.
What does MinMaxHR actually do?
It measures your hiring KPIs against a baseline, ranks applicants against the role through CandidRanker with the evidence attached, records every decision with a named owner and a reason, and gives leadership a governed view of hiring quality and speed.
What is CandidRanker?
CandidRanker is the candidate ranking and evaluation engine inside MinMaxHR. It evaluates applicants against the job description across eight dimensions and exposes the evidence behind the result. It is not a separate HAB product or ATS, and a named human keeps the final hiring decision.
How is a candidate scored?
Through CandidRanker, the ranking engine inside MinMaxHR: eight dimensions covering skills, tools, experience, education, certifications, projects, title similarity, and semantic similarity to the job description. Scoring is deterministic, so the same inputs always produce the same order.
How does MinMaxHR improve hiring governance?
Every score opens into its evidence, every decision has a named human owner and a written reason, and every change to weights, filters, skills, or terminology is logged in a scoring audit trail. That combination is what HR and procurement teams mean by governance.
Does the AI ever reject a candidate on its own?
Never. The system ranks and explains; a human shortlists, selects, rejects, or clears. We deliberately do not build automatic rejection, and we will say so in a procurement review.
Where does our data live?
In India (Google Cloud asia-south1), with malware scanning on upload, tenant isolation through row-level security, API keys held as SHA-256 hashes, and time-bounded read-only support access. Controls are built to SOC 2 alignment, with independent certification on the roadmap.
How quickly can we run a real pilot?
One role family, audited first and then piloted, with the go-live date agreed in scoping. Small enough to govern and real enough to measure against the baseline.
See how MinMaxHR turns a large applicant pool into a defensible shortlist.
Bring one active role and its applicant pool. We will show you the ranked evidence, the human decision gate, and where the current workflow is losing time.
No retainers to start · Pilot-first · Human-in-the-loop governance