Forensic OS for Hiring

The forensic operating system for modern hiring — AI shortlisting, fraud forensics, deepfake interview detection, and 130-point document verification in one console.

IntelliHire · Live Verdict Scanning
PR
Priya Raghavan
Sr. Backend · interview #4471
✓ REAL · 99.2%
MA
Marcus Chen
Sr. Backend · docs verified
✓ CLEAR
CA
Candidate 7F2A
3 fraud signals tripped
⚠ Suspicious
25%
Applicants Could Be Fake
7 in 10
Misrepresented Resumes
75%
Submit Fake Documents
20+ Hrs
Reviewing Fake Applications
The Problem

Hiring has an integrity problem.

Your ATS was built to move applicants through stages — not to tell you which of them are real. Every challenge below lives in a blind spot it was never designed to cover.

Risk 01

Synthetic candidates

AI face-swaps and deepfakes now pass live video interviews convincingly enough to fool a trained recruiter.

ATS: logs the candidate, not whether the face is real.
Risk 02

Proxy interviewers

A stronger stand-in quietly takes the technical screen for the person who actually applied for the role.

ATS: tracks the schedule, not who shows up.
Risk 03

AI-written applications

ChatGPT resumes, cover letters, and coached live answers blur who can genuinely do the work.

ATS: stores the text, can't tell it was generated.
Risk 04

Forged credentials

Fabricated degrees, experience letters, and payslips clear a basic eyeball review every day.

ATS: files the document, never verifies it.
Risk 05

Mass-apply bot fraud

Automated scripts and resume mills flood pipelines with junk applications and hidden risk.

ATS: counts every bot as a real applicant.
Risk 06

No shared fraud signal

Point tools don't talk to each other, so a red flag raised at apply is forgotten by the offer stage.

ATS: a system of record, not of trust.
Five Super Modules

One console. End-to-end hiring forensics.

Each module ships standalone or as part of the unified IntelliHire pipeline. Wire them into your existing ATS, video stack, and HRIS in under an hour.

01

AI Resume Shortlisting Agent

From thousands of resumes to a ranked shortlist in minutes.

Connect your ATS or drop a folder. Our agent reads every resume against the JD or your prompt, returns ranked matches with reasoning, and pushes the shortlist back to your pipeline.

  • Native ATS sync — Greenhouse, Lever, Workday
  • Prompt-tunable LLM ranker with explainable match reasoning
  • Bulk upload up to 500 resumes per batch, <2s per profile
Live Signal - Module 1
Priya Raghavan
92%
Marcus Chen
88%
Ana Lima
81%
Jordan Hale
64%
Live Signal - Module 2
VPN / Proxy72%
Disposable email48%
Geo anomaly61%
Device entropy88%
ALERT · candidate 7F2A · 3 signals tripped
02

AI Fraud Analyser

Digital footprint forensics for every shortlisted candidate.

Device fingerprint, IP reputation, email & phone enrichment, and social graph signals — fused into a single risk verdict. Catch proxy candidates and resume mills before the interview.

  • IP, VPN, datacenter & geo-anomaly checks
  • Email / phone breach + disposable detection
  • Device entropy & behavioural biometrics
03

AI Interview Deepfake Detector

Plugs into Zoom, Google Meet & Microsoft Teams.

An API and bot that joins the interview, scans each frame for face-swap artifacts, lip-sync drift, and AI-coached eye movement — then classifies the candidate as real, suspicious, or synthetic in real time.

  • Frame-level face-swap & GAN artifact detection
  • Eye-movement calibrator — detects AI prompt reading
  • Proxy / ghost-interviewer continuous detection
Live Signal - Module 3 Threat
Candidate Deepfake Scan
FACE-SWAP DETECTED
REC · 30fps
OFF-SCREEN GAZE
+
GAN
+
BLEND SEAM
SYNTHETIC · 97.3%
Face-swap probability97%
Lip-sync drift88%
GAN artifact91%
Gaze off-screen73%
99.4%
Accuracy on the DFDC
benchmark
Live Signal - Module 4
PDF
B.Tech_Certificate.pdf
2.1 MB · IIT Madras · 2018
1 Issue
  • EXIF metadata PASSED ✓
  • Error Level Analysis PASSED ✓
  • Font substitution FLAGGED ⚠
  • Issuer cross-verify PASSED ✓
124/130 Checks 95.4%
04

Tru. Doc

130-point document verification.

Tru.DOCS runs 130 forensic checks on every uploaded document — metadata anomalies, font substitution, pixel-level tamper traces, and issuer cross-validation. It flags edited, fabricated, or AI-generated docs in seconds.

  • 130-point check: EXIF, ELA, font, layout
  • Issuer cross-verification with 1,400+ entities
  • Detects AI-generated payslips & edited certificates
Full-Funnel Coverage

Fraud doesn't stop at the interview.

The modules share one signal layer, so a flag raised at application still counts at offer. These capabilities run across every stage of the funnel.

Identity gate

Pre-interview liveness plus passport scan before a candidate joins the call.

Continuous match

Session-long face continuity so a verified candidate can't be swapped.

Reconciliation

Reconciles written resume claims against what the candidate says live.

Enrichment

Validates identity and history against external data sources in real time.

Network flags

Catches VPNs, datacenter origins, and location spoofing instantly.

Contact forensics

Breach exposure, disposable detection, and metadata risk on contact info.

Bot honeypots

Hidden trap fields that quietly catch automated mass-apply scripts.

Audit trail

One signed, exportable verdict trail from application all the way to offer.

The Pipeline

Every candidate. Six forensic checkpoints.

A unified pipeline that connects ATS → shortlist → fraud → docs → deepfake → hire. One audit trail. One verdict.

01

Sourced

ATS sync + bulk upload

02

Shortlisted

LLM match vs JD

03

Fraud-screened

Device · IP · email

04

Docs verified

Tru.DOCS 130-point

05

Deepfake-cleared

Live interview probe

06

Hired

Audit trail signed

Built for regulated hiring.

IntelliHire is the only hiring intelligence platform that ships SOC 2 Type II, ISO 27001, GDPR / DPDP, and EEOC-aligned bias audits out of the box.

SOC 2 Type II
ISO 27001
GDPR
DPDP (India)
EEOC bias-audited
PII redaction