README.md

metadata

title: PII Engineer
emoji: 🏆
colorFrom: blue
colorTo: indigo
sdk: static
license: agpl-3.0
pinned: true
thumbnail: >-
  https://cdn-uploads.huggingface.co/production/uploads/65e5b479e6368eea3dc93535/WgnbhWYn9yi0rEoB2ZQIl.png
short_description: Fast, multilingual PII detection for privacy compliance.

PII Engineer

Fast, multilingual PII detection for privacy compliance (PDPA, PDPD, PDP Law, PIPL).

Detects 9 PII types across 13+ languages using fine-tuned GLiNER2 NER models with ONNX Runtime inference. Single binary, CPU-only, no GPU required.

Models

Model	Description
PII-Engineer-Multi-NER-v2.1	Multilingual NER (13+ languages, 9 PII types, ~280M params)
PII-Engineer-Chinese-NER-v1.0	Chinese-specific NER for improved CJK coverage

PII Types

person_name phone_number government_id street_address date_of_birth email_address passport_number license_plate bank_account_number

Languages

Primary: English, Malay, Tamil, Chinese, Indonesian, Vietnamese

Secondary: Thai, Hindi, Bengali, Korean, German, French, Russian

Links

• Live Demo
• GitHub

License

AGPL-3.0 — free for open-source use. Commercial license available at pii.engineer.