Writing
Long-form guides on AI audit logging, guardrails, PII and compliance — written while building SignalVault.
Start here
Where to put PII detection in an LLM app, which identifiers to validate rather than regex, how to rewrite the request before the provider sees it, and what to log instead of the value.
The EU AI Act's high-risk logging rules no longer start in August 2026. The new 2027 and 2028 dates, who they cover, and what to log in an LLM app.
A look at the features and infrastructure work we shipped in the first half of 2026 — from transparent proxy support and tool call audit logs to hardened auth and production reliability improvements.
Six months ago, logging LLM calls was enough. Now agents invoke tools, chain actions, and operate autonomously - and most audit logs miss the events that matter. Here's what the next version looks like.
Learn what AI audit logging is, what to log, encryption requirements, retention policies, and how audit logs enable SOC2/GDPR compliance.
A practical guide to SOC2 compliance for AI and LLM applications—controls, audit gaps, and how to build a compliance-ready AI stack.
Learn what AI guardrails are, how they differ from content moderation, and how to implement input validation, output filtering, and policy enforcement.
A practical guide to implementing guardrails for AI applications — detect PII, block secrets, and enforce policies before they become incidents.
Learn how to automatically detect and block personally identifiable information in LLM prompts before it reaches third-party AI providers.
What SOC2 and GDPR mean for AI applications, and how to build compliance into your AI stack from day one.