Integrate PrivacyPal's streaming Data Twin technology into your applications. Build compliant, multi-tenant AI solutions with real-time data anonymization.
import { PrivacyPalClient } from 'privacypal-sdk'; const client = new PrivacyPalClient({ apiKey: 'your-api-key', mode: 'streaming' }); // Stream sensitive data with Data Twins const stream = await client.createStream({ data: patientRecords, encoding: 'two-way' }); // Safe for RAG pipelines, AI processing await processWithAI(stream.anonymized);
Get up and running with PrivacyPal SDK in seconds
npm install privacypal-sdk
Add to your project with npm or yarn
Sign up for PrivacyPal Cloud access
Create privacy-safe data streams
Built for production workloads. Scale with confidence.
Build sophisticated AI solutions with privacy built-in from the ground up
Real-time Data Twin generation for continuous data streams. Perfect for RAG pipelines, vector databases, and AI training workflows.
Safely aggregate customer data for model training and analytics. Tell your customers "We don't train on your data" - because you're using Data Twins instead.
Meet HIPAA, GDPR, and SOC 2 requirements out of the box. Automated PII detection and anonymization for all major data types.
Deploy on your infrastructure with PrivacyPal Cloud. Complete control over sensitive data with data sovereignty guarantees.
Seamlessly integrate with LangChain, LlamaIndex, and popular RAG frameworks. Keep embeddings private while maintaining accuracy.
DSPM capabilities to discover and classify sensitive data in your internal databases, ensuring comprehensive data governance.
Real-world examples to get you started quickly
Create a simple data stream with automatic PII detection and anonymization
import { PrivacyPalClient } from 'privacypal-sdk'; const client = new PrivacyPalClient({ apiKey: process.env.PRIVACYPAL_API_KEY }); // Original sensitive data const records = [ { name: 'John Smith', email: 'john@example.com', ssn: '123-45-6789' }, { name: 'Jane Doe', email: 'jane@example.com', ssn: '987-65-4321' } ]; // Create anonymized stream const stream = await client.createStream(records); // Process with AI safely await sendToLLM(stream.data); // Data Twins used instead of real data
Protect embeddings and vector databases while maintaining semantic accuracy
import { PrivacyPalClient } from 'privacypal-sdk'; import { VectorStore } from '@pinecone/sdk'; const privacypal = new PrivacyPalClient({ apiKey: process.env.PRIVACYPAL_API_KEY, encoding: 'two-way' // Enable data recovery }); // Anonymize documents before embedding const documents = await loadSensitiveDocs(); const anonymized = await privacypal.anonymize(documents); // Store Data Twin embeddings safely await vectorStore.upsert(anonymized.embeddings); // Query and decode results const results = await vectorStore.query(userQuestion); const original = await privacypal.decode(results);
Aggregate multi-tenant customer data for model training without exposing sensitive information
import { PrivacyPalClient } from 'privacypal-sdk'; const privacypal = new PrivacyPalClient({ apiKey: process.env.PRIVACYPAL_API_KEY }); // Collect data from all your customers const customerAData = await fetchCustomerData('customer-a'); const customerBData = await fetchCustomerData('customer-b'); const customerCData = await fetchCustomerData('customer-c'); // Anonymize all customer data with Data Twins const anonymizedData = await privacypal.createStream([ ...customerAData, ...customerBData, ...customerCData ]); // Now safely train your AI model on aggregate data await trainModel(anonymizedData.data); // Tell your customers with confidence: // "We don't train AI on your real data" // Because you're using Data Twins! ✅