Veil is a self-hosted data privacy control plane: one detection & redaction engine across text, files, documents, databases, streams, object storage, telemetry gateways and AI/LLM calls — purpose-bound, with signed de-identification receipts.
Consent platforms hold the signal but aren't in the data path. Veil sits in the path — detects, redacts, tokenizes or masks in motion, checks the purpose it moves for, and hands back proof.
Every flow, stream, job, gateway call and MCP tool call declares the purpose it runs for and checks the data subject's consent before moving anything.
A portable, Ed25519-signed manifest per delivered artifact or call: what was detected, what policy acted, input/output hashes — verifiable offline by anyone.
Self-hosted in your cloud, with residency rules and a network egress boundary that refuses calls outside your allow-list.
Redaction API, playground, and a file studio: PDF, images (OCR), XLSX, CSV, text — redacted in the original format.
Teachable document types, flows that extract, enrich, look up and route documents from storage/mail/drives/SaaS to Kafka, JDBC, webhooks.
DB-to-DB masking, referential-integrity subsetting, and a re-identification risk gate on every export.
Kafka pipelines and S3/MinIO bucket jobs, redacted and checked for consent per message.
OTLP and Splunk HEC endpoints that scrub PII out of logs before they land downstream.
Wire-compatible with OpenAI/Anthropic SDKs — prompts tokenized, answers detokenized per audience, agent tool calls inspected.
RBA SystemsTexas · India