Security, Trust & ComplianceConference50min
Taming the Chaos: Security and Predictability in LLM-Powered Applications
The proposal presents practical patterns for building reliable LLM systems despite inherent randomness. It covers prompt-injection defenses, PII redaction, deterministic validation, self-correction, and multi-model consensus to improve safety and predictability. Examples use Java and LangChain4J, with concepts transferable to other languages and frameworks.
talk.summaryAiDisclaimer
Rustam MehmandarovMiles
talkDetail.whenAndWhere
Thursday, October 8, 13:50-14:40
TBA 3
talks.roomOccupancytalks.noOccupancyInfo
LLMs are probabilistic by nature. Ask the same question twice, get different answers. Great for creativity, terrible for business logic. How do you build reliable systems on top of something inherently unreliable?
Let's tackle this together – from security to predictability. We'll look at guardrails that block prompt injection and redact leaked PII before it leaves your system. Then we'll layer on deterministic validation that catches hallucinations with code, not hope. And when the model gets it wrong, we'll feed errors back for self-correction and use multi-model consensus to build confidence. Practical patterns, real code, no magic.
This isn't about making LLMs perfect – it's about building systems that fail safely and predictably when they're not.
The examples are in Java and LangChain4J; the concepts – transferable to other languages and frameworks.
Let's tackle this together – from security to predictability. We'll look at guardrails that block prompt injection and redact leaked PII before it leaves your system. Then we'll layer on deterministic validation that catches hallucinations with code, not hope. And when the model gets it wrong, we'll feed errors back for self-correction and use multi-model consensus to build confidence. Practical patterns, real code, no magic.
This isn't about making LLMs perfect – it's about building systems that fail safely and predictably when they're not.
The examples are in Java and LangChain4J; the concepts – transferable to other languages and frameworks.