v0.1.0 Stable Release

FHA Compliance
On The Edge.

The first latency-critical auditing engine powered by ModernBERT. Replace fragile regex with context-aware AI that runs locally on your infrastructure.

PyPI package coming soon! For now: pip install git+https://github.com/ZheWang-stack/FairProp-Inspector.git

MIT Licensed • Python 3.9+ • GPU Optional

user@fairprop-dev:~/inspector
~

01. Mission Control

FairProp Inspector was born out of a critical need in the real-estate AI era: Auditability without Compromise. As AI agents begin to handle property listings and tenant communications, the risk of subtle Fair Housing Act (FHA) violations has grown exponentially.

We provide the infrastructure to move compliance from slow human reviews to real-time, edge-native inference. By combining the linguistic understanding of ModernBERT with deterministic heuristic layers, we ensure your AI remains fair, compliant, and private.

Flash Attention

Built on ModernBERT topology. Supports 8192 context window with native Flash Attention 2.0 backend for O(N) complexity.

🔒

Privacy First

Zero data egress. Run full audit pipelines inside your VPC or directly on local hardware without calling external APIs.

🧠

Neuro-Symbolic

Combining statistical LLM strength with hard-coded FHA heuristics for 99.9% precision on edge cases.

02. Get Started in 3 Steps

From installation to your first compliance check in under 60 seconds.

1

Install

pip install git+https://github.com/ZheWang-stack/FairProp-Inspector.git

One command to install from GitHub

2

Run Example

python examples/quickstart.py

See it in action immediately

3

Customize

python src/trainer/train.py

Train on your own data

03. Ready-to-Run Examples

From simple demos to production-ready implementations.

04. Performance Metrics

Benchmarked on Intel i7-12700K CPU with ONNX Runtime optimization.

94%
Accuracy
vs 65% for regex rules
~18ms
P95 Latency
vs 800ms for cloud APIs
100%
Privacy
All processing local

05. Deployment Scenarios

Where FairProp Inspector makes the difference.

🤖 AI Agents

Integrate directly into LLM agent workflows to prevent non-compliant listing generation or discriminatory tenant responses in real-time.

📊 Mass Audits

Audit millions of historical property records locally at cost-zero without hitting rate limits or compromising sensitive data privacy.

Editor Widgets

Embed into real-estate CRM editors to give agents immediate feedback as they type listing descriptions, similar to Grammarly for FHA.

"

The convergence of privacy and performance is no longer a trade-off. FairProp enables us to deploy state-of-the-art compliance layers where data never leaves the edge.

The FairProp Philosophy

The FairProp Ecosystem

From individual listing checks to global portfolio mandates. One Standard. Any Scale.

YOU ARE HERE

FairProp Inspector

The edge-native inference engine.

ENTERPRISE LAYER

FairProp AI

The global compliance framework.

Multi-Jurisdictional (100+ Regions)
Portfolio-wide Analytics
Team Collaboration & Audit Logs
Visit Enterprise Platform