Shields AI
AI-Powered DNS Protection
Shields AI provides next-generation DNS security with machine learning-powered threat detection. It blocks malware, phishing, and trackers in real-time with sub-millisecond latency using eBPF kernel-level filtering. This is our flagship product with the clearest path to market. The SMB segment is underserved—Pi-hole requires technical setup, while Cisco Umbrella is enterprise-priced. Shields AI fills the gap with genuine AI differentiation and usage-based pricing.
How Shields AI Works
Shields AI filters DNS queries at the kernel level using eBPF, uses ML models for threat detection, and maintains <1ms latency while blocking threats in real-time.
Install Shield
Install Shields AI on your device or network. Works on Windows, macOS, Linux, routers, and Kubernetes clusters. One-click setup.
Configure DNS
Point your DNS to Shields AI (or use as default resolver). Works alongside existing DNS setups. No breaking changes to your network.
Real-Time Filtering
Every DNS query is processed in real-time. eBPF kernel-level filtering ensures <1ms latency. Queries checked against blocklists and ML models instantly.
Threat Detection
ML models analyze queries for malware, phishing, DGA (Domain Generation Algorithms), and suspicious patterns. Blocks threats before they reach your devices.
Custom Rules
Create custom blocklists, whitelist trusted domains, set time-based rules, and configure family profiles. Full control over your network security.
Monitor & Analyze
Real-time dashboard shows blocked threats, query statistics, privacy scores, and network health. Export logs for compliance and analysis.
Use Cases
Home Network Protection
Protect all devices on your home network. Block malware, phishing, and trackers. Parental controls for safe browsing.
Business Network Security
Enterprise-grade DNS security for businesses. Protect entire networks from threats, compliance reporting, centralized management.
Privacy Enhancement
Block trackers, ads, and data harvesting. Privacy scoring shows which domains track you. Minimize your digital footprint.
Development & Testing
Block malicious domains in development environments. Prevent accidental connections to dangerous sites during testing.
Key Features
Sub-Millisecond Latency
eBPF kernel-level filtering processes DNS queries in <1ms. No noticeable impact on browsing speed. Handles 100K+ queries per second.
ML-Powered Threat Detection
Machine learning models detect malware, phishing, DGA domains, and behavioral anomalies. Continuously learning and improving.
Customizable Blocklists
Built-in blocklists plus custom lists. Import from community sources, create your own, whitelist trusted domains.
Privacy Scoring
Privacy score algorithm analyzes domains for tracking behavior. See which sites collect data and how aggressively.
Family Profiles
Create profiles for different family members. Age-appropriate filtering, time-based rules, and usage reports.
Real-Time Dashboard
Monitor blocked threats, query statistics, network health, and privacy scores in real-time. Beautiful, intuitive interface.
DNS-over-HTTPS (DoH)
Encrypt DNS queries using DoH. Prevent ISP snooping and DNS hijacking. Enhanced privacy and security.
Kubernetes Integration
Deploy as Kubernetes service for container environments. Protect microservices, APIs, and containerized applications.
Pricing Plans
Free Tier
- Basic DNS filtering
- Up to 5 devices
- Standard blocklists
- Basic dashboard
- +2 more features
Home Subscription
- All free features
- Unlimited devices
- Advanced AI threat detection
- Custom blocklists
- +4 more features
Family Subscription
- All Home features
- Multiple family profiles
- Advanced parental controls
- Usage reports
- +3 more features
Enterprise
- All Family features
- Unlimited devices/users
- Advanced threat intelligence
- Custom deployment
- +4 more features
Built with AI
Shields AI was developed through human-AI collaboration. Core components are open source (MIT licensed).
Open Source Core
Core engines and libraries are MIT licensed. Audit, contribute, or self-host.
Premium Features
Enterprise features, premium UI, and dedicated support available commercially.
Self-Hosted
Deploy on your infrastructure with Docker or Kubernetes. Full control, full privacy.
Development Story
Incubation Timeline
Research & Design (August-December 2025)
Key Achievements
- Benchmarked existing DNS solutions (50+ performance tests)
- Designed eBPF kernel-level filtering architecture
- Planned lightweight ML models (<1ms inference)
- Created high-performance blocklist system
- Designed zero-trust DNS architecture
- Researched sub-millisecond optimization strategies
Technical Architecture
DNS Filtering Architecture
eBPF Kernel Filtering
Kernel-level packet processing for <1ms latency. Zero-copy packet inspection, lock-free data structures, per-CPU cache design.
ML Threat Detection
TensorFlow Lite models for DGA detection, behavioral anomaly detection, phishing pattern recognition. <0.1ms inference latency per query.
Blocklist System
Memory-mapped blocklists for instant lookup, 50K+ entry LRU cache, 8+ built-in blocklist sources, automatic updates every 4 hours.
Rust Core Engine
Hickory DNS library for resolution, tokio async runtime, DashMap for concurrent lookups, AHash for fast hashing.
Dashboard & API
Next.js dashboard with WebSocket real-time updates, RESTful API with OpenAPI docs, Prometheus metrics export, GraphQL query interface.
Privacy Scoring Engine
50+ privacy signals analyzed per domain, tracker categorization (analytics, ads, social, etc.), data flow mapping visualization.
Deployment Options
Docker containers, Kubernetes operators, systemd services, router firmware (OpenWrt), home automation (Home Assistant).
Vector Database
Qdrant for threat intelligence storage, vector embeddings for similarity search, real-time threat correlation engine.