Skip to main content
Security & Trust

Built for enterprise scale and security

We compress your prompts and route them efficiently, but we never compromise your data. Laghav is designed from the ground up to ensure your prompt content remains yours.

Our Core Promise: No Prompt Storage

Laghav processes your prompts in-memory and discards them immediately. We never store, log, or train on your prompt content. We only retain metadata (token counts, costs, models used) for your dashboard analytics and billing.

Data Encryption

All data is encrypted in transit using TLS 1.3 and at rest using AES-256. API keys are hashed with HMAC-SHA256 before storage — we can never read your raw API keys.

Multi-Tenant Isolation

Customer data is strictly isolated using Row-Level Security (RLS) in PostgreSQL and tenant-namespaced keys in Redis. Your data is invisible to other tenants.

PII Masking

Available on Scale plans, our PII masking engine (powered by Microsoft Presidio) redacts sensitive information before the prompt leaves your perimeter, ensuring compliance with DPDP and GDPR.

Compliance

Laghav complies with the Indian Digital Personal Data Protection Act (DPDP) and EU GDPR. We offer standard Data Processing Agreements (DPA) for all Business and Enterprise customers.

Infrastructure Security

  • Hosted on AWS ap-south-1 (Mumbai) ensuring data localization for Indian customers.
  • Protected by Cloudflare WAF and DDoS mitigation at the edge.
  • Regular vulnerability scanning of all dependencies via Snyk.
  • Strict least-privilege IAM roles for all services and databases.

Authentication & Access Control

  • Passwordless authentication via magic links only.
  • Granular API keys scoped by application.
  • Role-based access control (Admin vs Developer) for team members.
  • Audit logs of all key creation, rotation, and deletion events.

Responsible Disclosure

We take security seriously. If you believe you have found a security vulnerability in Laghav, please report it to us immediately. We will investigate all legitimate reports and do our best to quickly fix the problem.

Email your findings to: security@laghav.ai