In-Person

The strongest verification mode. Combines a NIN lookup with a biometric face match against a freshly-captured selfie of the holder.

POST/verify-nin/in-person

Request

Field	Type	Required	Notes
nin	string	✅	Exactly 11 digits
selfie_image	string	✅	Base64-encoded JPEG of a live selfie (no data:image/jpeg;base64, prefix)
reason	string	✅	A key from /nimc-reasons

Tip

The selfie must be a fresh capture of the holder taken at the moment of verification. Photo-of-a-photo and other spoof attempts are rejected by NIMC’s match engine.

curl

# Base64-encode the JPEG without newlines
SELFIE_B64=$(base64 < selfie.jpg | tr -d '\n')

curl -X POST https://api.randaverify.com/verify-nin/in-person \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d "$(jq -n --arg nin '12345678901' --arg img "$SELFIE_B64" --arg reason 'corporate' \
       '{nin:$nin, selfie_image:$img, reason:$reason}')"

Python

import base64, requests

with open("selfie.jpg", "rb") as f:
    selfie_b64 = base64.b64encode(f.read()).decode()

resp = requests.post(
    "https://api.randaverify.com/verify-nin/in-person",
    headers={"Authorization": f"Bearer {token}"},
    json={
        "nin": "12345678901",
        "selfie_image": selfie_b64,
        "reason": "corporate",
    },
    timeout=45,  # face match adds latency
)

Response (200)

Returns the standard normalised record. NIMC’s match score is reflected in the underlying response and surfaced via the data.status / data.message fields when relevant — a 200 from this endpoint already implies the match succeeded above NIMC’s threshold.

Failure modes

Status	When	Recover by
400	NIN invalid, selfie unreadable, or match score below threshold	Re-capture the selfie in better lighting; ensure the NIN matches the person in the photo.
402	Org wallet empty	Top up.
500	Upstream face-match service unreachable	Retry with backoff.

Cost & idempotency

Charged 1 unit on success, same as standard verification. A failed match (below threshold) still costs 1 unit because NIMC ran the comparison — see Wallet & units.