KYC review that runs in seconds, not days
Read identity documents, pull Aadhaar and PAN fields to JSON, mask PII before storage, and validate with confidence scores — all over a flat per-page REST API built for regulated workflows.
Manual KYC is slow, expensive, and error-prone
Onboarding teams retype Aadhaar and PAN numbers by hand, eyeball blurry scans, and copy fields into core systems one document at a time. Every manual touch adds latency and risk.
Drop-off from delay
Applicants abandon onboarding while documents sit in a manual review queue. Hours of turnaround translate directly into lost conversions.
Transcription errors
A single mistyped Aadhaar or PAN digit fails downstream validation and forces a re-review. Manual keying has no confidence signal to flag the risky fields.
Raw PII everywhere
Full identity scans get stored in ticketing tools and shared drives unmasked, widening your exposure surface long before any breach.
A clean pipeline from scan to validated record
Chain four hosted APIs into one KYC flow — each pay-as-you-go, each returning structured JSON with bounding boxes and confidence.
Read every identity document, in any condition
OCR Full-Text turns scanned Aadhaar, PAN, passports, and address proofs into machine-readable text and a searchable PDF. Built-in deskew, denoise, and orientation correction handle the photos people actually upload.
- Inputs PDF, TIFF, JPG, PNG, and scanned images, multi-page
- Multi-language including Hindi, Tamil, Telugu, and Arabic
- Rs 0.25 /page (about $0.003 / page)
Extract the fields you need as JSON
Extract Basic and Extract Pro pull named fields — Aadhaar number, PAN, name, date of birth, address — from KYC forms and ID documents and return them as structured JSON with bounding boxes, so you can write straight to your onboarding system.
- Extract Basic: structured fields + bounding boxes — Rs 0.90 /page
- Extract Pro: adds confidence scores, guardrails, and validation rules — Rs 1.40 /page
- Synchronous for small docs, or asynchronous with a webhook callback
Mask PII before anything is stored
PII Masking redacts Aadhaar, PAN, account numbers, names, addresses, phone numbers, and signatures while preserving the original layout. Store the masked copy in your tools and keep raw identity data out of downstream systems.
- Configurable mask rules per document type
- Output preserves the original document layout
- Rs 0.25 /page (about $0.003 / page)
Validate with confidence scores and guardrails
Extract Pro returns a confidence score per field plus configurable validation rules, so your pipeline can auto-approve clean records and route only the uncertain ones to a human reviewer. Built for fintech and regulated workflows.
- Per-field confidence to drive straight-through processing
- Validation rules and guardrails on complex documents
- Higher accuracy where it matters most for KYC
The APIs behind a KYC flow
Mix and match. Flat per-page pricing, no surge pricing and no tiered overages.
OCR Full-Text
Full text extraction and a searchable PDF from scanned identity documents, with auto-deskew and orientation correction. Rs 0.25 /page.
Explore OCR →PII Masking
Redact Aadhaar, PAN, account numbers, names, and signatures while preserving layout, before documents land in storage. Rs 0.25 /page.
Explore Masking →Extract Pro
Structured fields plus confidence scores, guardrails, and validation rules for higher accuracy on regulated KYC docs. Rs 1.40 /page.
Explore Extract Pro →Built with data protection in mind
Process identity documents without expanding your retention footprint.
Documents are auto-purged
Documents are automatically purged after processing. We do not retain your documents and we do not train models on customer data. The API returns the extracted JSON and the processed output, and the source file is gone.
DPDP-aware design
Masking and auto-purge let you minimise the personal data you store, supporting data-minimisation under India's Digital Personal Data Protection framework. Encryption is TLS 1.3 in transit and AES-256 at rest.
Regional processing
Indian traffic is processed and hosted in India (GCP Mumbai); Africa and Middle East traffic is processed in Johannesburg. Need a specific data-residency region or high-volume pricing? Talk to our team →