Nex Systems

The technology behind every check.

Nex 360 combines document reading, biometrics, risk signals, entity matching, detection rules and AI-assisted analysis. Each one is built to keep evidence in front of the people who decide.

Book a demo See how we use AI

Documents

Read the document. Check the details.

Turn passport and ID images into structured data, then test it against what the document should contain.

Explore document verification

MRZ check

P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<
L898902C36UTO7408122F1204159ZE184226B<<<<<10
  • Document numbercheck digit valid
  • Date of birthcheck digit valid
  • Expirycheck digit valid
Example using the ICAO specimen passport.
Field extraction
AI-assisted reading turns printed fields into structured details.
Machine-readable zone check
Machine-readable zones are parsed and every check digit recalculated.
Barcode and QR decoding
PDF417 and QR codes are decoded where the document carries them.
Document definitions
Passports, ID cards, driving licences, residence permits and voter IDs, described by country and type.
Consistency checks
Expiry, number formats and printed-versus-MRZ disagreements are flagged for review.
Capture inspection
Visible capture problems, alteration signals and type mismatches are surfaced for review.

Biometrics

Connect the person to the document.

Face comparison
Compare the applicant's selfie with the portrait on their document.
Liveness evidence
Confirm a live capture, as a check separate from the face match.
Face comparison Live presence check

Risk signals

See the context around each attempt.

Device and network
Device changes, approximate location and VPN, proxy or Tor indicators.
Blocklists
Stop identities, IP addresses or faces your team has already flagged.
Device & network signals

Screening

Match names the way people write them.

Sanctions, PEP and adverse-media screening that explains why each result appeared.

Explore sanctions & PEP screening
Fuzzy and phonetic matching
Spelling variants and sound-alike names still reach the right record.
Ownership exposure
Ownership chains are tested against each sanctions regime's own blocking rule.
Source evidence
Every finding names its source and when the copy was taken.
Adverse-media analysis
Reports are assessed for who they concern and the stage the case has reached.
Change detection
Monitoring flags material changes, not cosmetic refreshes.

Detection

Write the rule. Test it first.

Rules and risk methods built around your programme, not a fixed black box.

Explore transaction monitoring 
when count(transactions, 7d) > 20
and sum(amount, 7d) > 50,000
then raise alert "Rapid movement"
Backtest · last 90 days
Example rule, simplified for illustration.
Rule language
Thresholds, rolling counts and sums, and conditions across related subjects.
Backtesting
Run a saved rule over past transactions before it goes live.
Risk methodology
Ratings built from weighted factors your team can explain.

Security

Protection in the platform itself.

  • Field-level encryption

    Sensitive verification data is encrypted with keys we rotate.

  • Access by role

    People see and change only what their role allows.

  • Audit history

    Actions are recorded, with integrity checks on the history.

  • Signed integrations

    Versioned APIs, scoped keys and signed webhooks.

Read about security & data protection

Questions about our technology

No. Checks test the data and the visible signals in an image. They give reviewers evidence about what needs attention. A person still decides whether to accept the document and the applicant.

See the technology on your own cases.

Book a demo Contact us