AI Governance for
Companies That Can't Afford
to Get It Wrong
Sentorix sits between your apps and AI providers. Every request is scanned for PII, checked against your policies, and logged immutably — in under 20ms.
Trusted by teams building AI-powered fintech, healthcare, and HR platforms
import openai
client = openai.OpenAI(
api_key="sk-..."
)import openai
client = openai.OpenAI(
api_key="sk-...",
base_url="https://api.sentorix.io/v1/gateway"
)
# PII detected. Policies enforced.
# Audit logged.Your team is already using AI.
You just don't know how.
Most companies adopt AI tools faster than they can govern them. The result is legal liability, compliance risk, and reputational damage hiding in plain sight.
Sensitive data in AI prompts
Employees paste customer PII, financial records, and health data directly into ChatGPT and other AI tools every day.
Zero visibility into AI usage
When regulators ask what data was sent to AI providers and when, most companies have no answer. That silence is expensive.
AI systems under attack
Malicious users craft prompts to manipulate your AI features, extract system prompts, or bypass your application logic.
Unauthorised AI tool usage
Your policies say no external AI. Your employees use it anyway. You find out when something goes wrong.
How Sentorix works
Every AI request flows through the Sentorix gateway before reaching your LLM provider.
Inspect
Every request is scanned for PII (names, emails, SSNs, financial data) and prompt injection attempts before reaching your LLM.
Enforce
Your policies are applied. High-risk requests are blocked. PII is redacted. Every decision is made in under 20ms — invisible to your users.
Log
Every interaction is hashed and written to tamper-proof audit storage. 7-year retention. Compliance reports in one click.
Everything your compliance team has been asking for
Live in 5 minutes
Change your base URL. That's it. Sentorix is OpenAI-API compatible — no SDK changes, no code refactoring, no new infrastructure to manage.
- OpenAI compatible
- Anthropic compatible
- AWS Bedrock compatible
- Works with any language or framework
Real-time PII detection
Powered by Microsoft Presidio. Detects and redacts 12 entity types before they reach your AI provider. Configurable thresholds per tenant.
- Names, emails, phone numbers
- Credit cards, SSNs, IBANs
- Medical records, IP addresses
- Custom entity types coming soon
Audit-ready from day one
Immutable audit logs stored with AWS S3 Object Lock in COMPLIANCE mode. 7-year retention. Export for regulatory submissions in one click.
- EU AI Act ready
- HIPAA compatible
- India DPDP Act
- SOC 2 evidence
Built for regulated industries
High-risk AI systems require documentation, human oversight, and audit trails. Sentorix makes compliance automatic.
Healthcare AI applications cannot send PHI to external AI providers without explicit controls. Sentorix detects and blocks PHI automatically.
Personal data processing requires purpose limitation and audit trails. Sentorix enforces both at the AI layer.
Data minimisation and processing records are requirements, not optional. Sentorix creates them automatically.
If Cloudflare protects your web traffic,
Sentorix protects your AI usage.
| Cloudflare | Sentorix | |
|---|---|---|
| Protects | Web traffic | AI traffic |
| Inspects | HTTP/HTTPS | LLM prompts |
| Blocks | Bad actors | PII + attacks |
| Logs | Access logs | Audit trail |
| Sits between | You + web | You + AI |
Built by developers, for developers
Sentorix is designed to be invisible. Your developers keep using the OpenAI SDK they know. Governance happens automatically.
- OpenAI-compatible API (drop-in replacement)
- Python and Node.js SDKs (coming soon)
- Streaming response support
- Per-tenant policy configuration
- Webhook notifications for violations
- 99.9% uptime SLA (coming soon)
- Sub-20ms overhead (p99)
- SOC 2 Type II (in progress)
import openai
# Before: standard OpenAI client
# client = openai.OpenAI(api_key="sk-...")
# After: add Sentorix in one line
client = openai.OpenAI(
api_key="sk-your-openai-key",
base_url="https://api.sentorix.io/v1/gateway",
default_headers={
"X-Sentorix-API-Key": "your-sentorix-key",
"X-Tenant-ID": "your-tenant-id",
"X-User-ID": "user-123"
}
)
response = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": prompt}]
)
# PII detected. Policies enforced. Audit logged.See Sentorix in action
Request a personalised demo. We'll show you how Sentorix integrates with your stack and protects your AI usage in under 15 minutes.
What you get in a demo
- Live PII detection on your actual prompts
- Policy configuration walkthrough
- Audit log and compliance report demo
- Integration guide for your tech stack
- Answers to your compliance questions
- Pricing tailored to your usage
No sales pressure. No commitment.
Response within 24 hours.