AI/ML●●Solid
I built proxy that keeps RAG working while hiding PII
Consistent pseudonymization beats redaction when RAG embeddings must survive.
Big BrainSolve My Problem
rohansx
402mo ago
Back to browse
A proxy to hide PII information from LLM requests
by guimaster97·Apr 6, 2026·1 point·0 comments
AI Analysis
●MidSolve My Problem
Yet another PII redaction proxy when Lakera and Portkey already dominate this space.
Strengths
- •Sub-50ms edge processing is genuinely fast for compliance middleware.
- •Audit logs designed specifically for SOC2 audit requirements.
Weaknesses
- •Regex plus NER detection is standard approach used by established competitors.
- •Enterprise security teams already have budget for Lakera, Portkey, or custom solutions.
Category
Target Audience
Enterprise engineering teams blocked by legal/compliance
Similar To
Lakera Guard · Portkey · PromptLayer
Post Description
Hi HN. I'm exhausted from seeing legal departments block the use of LLMs because devs were accidentally sending API keys and customer data to OpenAI. Enterprise solutions cost a fortune and take months to integrate. So I built Shieldnod. It's middleware running at the edge that uses regex and NER rules to censor [NAME] or [STRIPE_KEY] in under 50ms before hitting the LLM. I want to test if anyone else is experiencing this. Destroy my landing page or tell me why this approach is idiotic.
Similar Projects
Infrastructure●Mid
Sentinel – Go LLM Proxy with 13ms Semantic Cache and PII Scrubbing
Multi-model LLM router with semantic cache, but caching+fallback already exist (Anthropic, LangSmith, Unify).
SlickCrowd Pleaser
ChipShotz
113mo ago
Security●●Solid
Aegis.rs, the first open source Rust-based LLM security proxy
Zero-code LLM firewall; heuristics under 1ms, optional Groq semantic layer.
Solve My ProblemSlick
ParzivalHack
223mo ago
Security●●●Banger
OpenGuard
Drop-in LLM traffic guard with PII redaction and prompt injection detection, one command.
Solve My ProblemSlick
everlier
102mo ago
Security●●●Banger
Veil a Drop-in PII redaction proxy for any LLM API
Stops zero-width Unicode bypasses that break standard PII filters before LLM calls.
WizardrySolve My Problem
A5omic
201mo ago
Security●●●Banger
OnGarde – Runtime content security proxy for self-hosted AI agents
Transparent proxy blocks agent leaks to LLM APIs without touching agent code.
Solve My ProblemSlick
antimaterial
213mo ago