Wenqi Glantz

Background

Writes detailed, implementation-level walkthroughs of LLM application components, including a head-to-head evaluation of the two mainstream guardrail frameworks.

Articles in This Vault

• NeMo Guardrails, the Ultimate Open-Source LLM Security Toolkit

Key Contributions

• Benchmarked NeMo Guardrails against Llama Guard — equal 89% input-moderation accuracy on 18 adversarial prompts, at materially different hardware cost

Concepts

• LLM Guardrails · RAG · Hallucination Detection

Tools

• LlamaIndex