Setup and Handoff
We stand up PII redaction in your own cloud, configure and validate it on your data, then hand you a running system you own. No in-house engineering team required.
Self-hosted PII redaction
Your Cloud. Your Data.
Open source, self-hosted PII and PHI redaction software that runs entirely inside your cloud, built for healthcare, finance, legal, and government workloads.
In production since 2017. Book a 30-min review or explore the toolkit →.
Trusted in
- Healthcare
- Finance
- Legal
- AI & Machine Learning
- Government
- Contact Centers
- Insurance
- AI Training Data
- Education
Consulting
Your trusted PII redaction partner
No in-house engineers? Bring in the team that built the software. We design and deploy PII redaction inside your own cloud, validate it on your data, and hand you a system you own outright. You work directly with the people who wrote the code, not a vendor you renew every year.
Privacy Architecture
We design end-to-end PII protection for your data and AI workloads: data flows, redaction layers, audit trails, and the guardrails that keep generative-AI features aligned with HIPAA, GDPR, and CCPA.
Custom Detection Models
Off-the-shelf models miss the identifiers that matter most in your domain. We train specialized PII and PHI detectors on your data, measured against precision and recall you can put in front of an auditor.
See all consulting services, case studies, and the solution brief →
Which open source Philterd tool do I need?
Every tool in the Philterd toolkit is open source and free to run. Start from what you're trying to do, and each path maps to the tools that solve it.
I want to redact PII
Pick the surface that matches where the text lives: an API, an embedded library, or an LLM gateway.
- Philter → Self-hosted redaction API. Drops into any HTTP-based pipeline.
- Phileas → The same engine as a library in Java, Python, and .NET. No service hop.
- Philter AI Proxy → Drop-in proxy for OpenAI and Anthropic. One URL change.
I want to find and watch PII
Map where PII already lives, then watch how it moves through the systems you care about.
I want to author, review, and measure policies
Build the policy, override what the system gets wrong, and measure how well it works over time.
- Redaction Policy Editor → Visual no-code builder for compliance teams.
- PhiSQL → Declarative DSL. Policy-as-code for engineering teams.
- Arbiter → Human-in-the-loop review with structured exemption codes.
- Philter Scope → Benchmarks precision, recall, and F1 so you can tune with numbers.
Why teams choose Philterd
Three principles shape everything we build: your data never leaves your perimeter, the engine is open source and auditable, and the models are purpose-built for PII and PHI.
Data Sovereignty
Philter and the rest of the Philterd toolkit run inside your cloud. Your data never leaves your perimeter, never reaches a third-party API, and never lands in someone else's logs.
Open Source Integrity
Transparency is the only way to verify privacy software. Our core engine is Apache 2.0 licensed, so your engineers can read every line, audit every decision, and extend the stack on their own terms.
Purpose-Built AI
Generic LLMs make poor privacy filters. We train and ship specialized NLP and deep-learning models built specifically for PII and PHI detection. They are accurate, tunable, and operationally affordable at scale.
Compliance and Trust
Philterd provides a zero-trust architecture designed to support your HIPAA, GDPR, and CCPA compliance efforts. The discovery engine operates entirely within your infrastructure: 100% data sovereignty, no external API dependencies, no third-party data training. Detection uses NLP and is probabilistic, so validate coverage against your own data; because you self-host, you remain the data controller responsible for the output.
To support HIPAA Safe Harbor de-identification, we pair high-speed pattern matching for structured identifiers with specialized AI models for everything else, with detection and handling strategies for all 18 protected identifier categories under 45 CFR § 164.514. Healthcare and life-sciences organizations can automate much of the de-identification work across massive datasets while preserving the utility the data needs for research and innovation. Validate coverage against your own data before relying on it.
PII redaction software: frequently asked questions
What is PII redaction software?
PII redaction software finds and removes or replaces personally identifiable information (PII), and protected health information (PHI), in text so the sensitive values are not exposed. Philterd's redaction software is self-hosted and open source, so it runs inside your own environment instead of a third-party service.
Is Philterd's PII redaction software free and open source?
Yes. The toolkit is released under the permissive Apache License 2.0, free to run, and developed in the open on GitHub.
Does it run self-hosted, without sending data to a third party?
Yes. It runs entirely inside your own cloud, VPC, data center, or an air-gapped network, so sensitive text never leaves your boundary to be redacted.
Can it support HIPAA, GDPR, and CCPA compliance?
It is designed to support HIPAA, GDPR, and CCPA compliance efforts by removing identifiers before text is shared or stored. Detection is probabilistic, so you remain responsible for validating the output against your own data.
How accurate is the PII detection?
Detection uses configurable policies and trained models rather than regex alone, which helps reduce how much sensitive data is exposed. No detector catches every instance, so tuning and validating against your own data is recommended.
How do I deploy the PII redaction software?
Launch Philter from the AWS, Google Cloud, or Azure marketplaces, or self-host it from GitHub and call its REST API. See the open source toolkit to get started.
Three ways to get started
Same redaction engine, three paths. Pick the one that fits your team.
Free forever
Open Source
$0 · Open source
Run the entire Philterd toolkit yourself. Full source on GitHub. No license keys, no usage caps, no commercial review.
Per-hour billing
Philter on Cloud Marketplaces
From $0.49/hr · ~$360/mo
Deploy Philter (our turnkey redaction API) into your VPC from the AWS, Google Cloud, or Azure marketplace. Production-ready in minutes; billed through your existing cloud account. The other Philterd tools are not yet on the cloud marketplaces.
Deploy in 5 minutes
Available on AWS, Google Cloud, and Azure. AWS Marketplace list price shown. See the full TCO comparison →
Engagement-based
Engaged
Request a quote
Work directly with the people who built the toolkit. Custom NLP models, privacy architecture, embedded engineering, and production deployment with full handoff.
From the blog
Practical posts on PII redaction, AI privacy, and self-hosted compliance.
· AI, Models, PhEye
How to Tell if a PII Redaction Model Is Any Good
A model count says nothing about quality. What proves a PII redaction model works: held-out evaluation, precision and recall, and an auditable model card.
Read post →· Philter, Redaction
Precision, Recall, or F1: Which Redaction Metric Matters Most in Your Industry
The right redaction metric depends on your industry. See how healthcare, legal, finance, marketing, and research each prioritize precision, recall, or F1.
Read post →· Philter, Redaction
Redact PII From Your AI Agent in One Block of Config
Philter MCP exposes PII and PHI redaction as Model Context Protocol tools that Claude Desktop, Claude Code, Cursor, and other MCP clients call mid-conversation.
Read post →