In this episode of AI That Works, we dive deep into the often-overlooked but crucial topic of PII (Personally Identifiable Information) redaction! We whiteboard out the different types of PII and how to handle them appropriately, because leaking sensitive data is a no-go.
We discuss how to build security controls directly into your codebase, even before you get to the AI layer. Think RBAC and other deterministic methods to ensure data protection. Why make your LLM struggle with tasks it's not designed for? Let's optimize for speed, cost, and reliability by handling PII redaction strategically.
Check out our github: https://www.github.com/boundaryml/baml
AI That Works repo: https://github.com/ai-that-works/ai-that-works
Socials:
X: https://x.com/boundaryml
Discord: https://discord.com/invite/yzaTpQ3tdT
LinkedIn: https://www.linkedin.com/company/boundaryml/
⏱️ Chapters:
Intro
AI Problem Misconceptions
PII Redaction and LLMs
Writing Code Examples
Secure Data Sets
Local Models Improving
Context Engine
Major Enterprises Cloud Code
Loop Running in Prod
Transcript
I've watched this one closely. Ask me anything about it and I'll point you to the exact moment.