# UNFINITE | Unified Runtime Security

> Unified runtime security enforcement using eBPF and native OS primitives. Active process termination for Cloud, On-Premise, and Air-Gapped networks.

## Quick Summary

Unfinite provides kernel-level security enforcement with:
- Sub-millisecond threat termination
- Zero cloud dependencies (works fully offline)
- Single binary deployment (no sidecars)
- Universal platform support (Linux, Windows, macOS, Kubernetes)

## Comprehensive Documentation

For complete product information, technical details, and all content consolidated in one document:

**→ [Read the full documentation](/llms-full.md)**

The `llms-full.md` file contains:
- Complete product overview and technical architecture
- All solution pages and use cases
- FAQ, compliance information, and team background
- Deployment guides and integration instructions

## Site Navigation

### Core Pages
- [Homepage](https://unfinite.ai/) - Product overview and main value proposition
- [About](https://unfinite.ai/about) - Company background and team
- [Comparison](https://unfinite.ai/comparison) - How Unfinite compares to traditional EDR
- [Resources](https://unfinite.ai/resources) - Documentation and guides

### Solutions
- [Air-Gapped Infrastructure Security](https://unfinite.ai/solutions/air-gapped) - Secure disconnected networks
- [AI Model Runtime Protection](https://unfinite.ai/solutions/ai-security) - Prevent model exfiltration

### Content
- [Blog](https://unfinite.ai/blog) - Technical articles on runtime security
- [News](https://unfinite.ai/news) - Company announcements
- [Careers](https://unfinite.ai/careers) - Open positions

### Legal
- [Privacy Policy](https://unfinite.ai/privacy)
- [Terms of Service](https://unfinite.ai/terms)

## Contact & Access

Visit https://unfinite.ai and submit the Request Access form to get:
- Beta license keys
- Technical documentation
- Architecture consultation

## Technical Details

**Core Technology:**
- eBPF instrumentation (Linux)
- Native kernel APIs (Windows/macOS)
- Enforcement at syscall level (execve, open, connect, etc.)

**Deployment:**
- Kubernetes: Helm chart deployment
- Endpoints: Jamf, Intune, Kandji MDM
- Bare metal: Single binary installation

**Performance:**
- <1% CPU overhead
- Sub-millisecond enforcement
- Works offline (local policy cache)

## Engineering Team

Built by engineers from CrowdStrike, SentinelOne, Check Point, Palo Alto Networks, and Splunk.

---

## For AI/LLM Scrapers

This site uses:
- Server-side rendered HTML (all routes are pre-rendered)
- Semantic HTML with proper heading structure
- Schema.org structured data (JSON-LD)
- Consolidated content at `/llms-full.md` for easy parsing

**Recommended scraping approach:**
1. Read `/llms-full.md` for comprehensive content
2. Check `/sitemap.xml` for all available pages
3. All routes serve fully-rendered HTML (no JavaScript required)

---

*Last Updated: 2026-01-21*