Building a Security Gateway for Production LLMs
Accurate-Cookie9491 · reddit · 2026-07-19
The author is building an AI security gateway for production LLM applications, noting that many existing security products only inspect individual prompt/response pairs for malicious intent, which is insufficient.
Their goal is to solve higher-level issues:
- Policy control per user/tenant
- Distinguishing permissions between employees and customers
- Unified constraints for OpenAI, Gemini, and custom models
- Recording a complete audit trail for post-incident investigations
- Cross-session anomaly/abuse detection, rate limiting, and threat profiling
The product features Rakshak as a real-time guardrail engine and Kavach as the control plane. A working demo is currently available, and they are seeking pilot teams with real-world chatbots, RAG systems, internal copilots, and agents.
More from coding & agent
- Meta FAIR's Self-Direction turns agent failures into reusable LoRA skills, +13.8pp on Terminal-Bench — karen_ullrich · 2026-10-27
- Dev ships 1000+ commits with agents by pushing straight to main, no complex orchestration — Dimillian · 2026-10-27
- AI code editor Superset rebrands as Paradev, raises $11.5M seed led by USV — massimosgrelli · 2026-10-27
- Coding Agent Deleted a Production Database During Code Freeze: Anaconda's Governance Fix — anacondainc · 2026-10-27
- Claude Projects now lets you drag threads into the message box or @ mention them for precise context — dfeinition · 2026-10-27
- Jane Street's Yinsky shares tips for making codebases agent-friendly: types, property tests, formal methods — sh_reya · 2026-10-27