Are you experiencing performance issues with your Graylog instance? Are late-night alerts and unexplained slowdowns keeping you up at night? You're not alone if you’re dealing with license limit violations without a clear cause. In this session, we’ll share our experiences with these common Graylog challenges and the practical solutions we’ve developed to overcome them.

So you’ve gone ahead and ingested every log you can think of and built a plethora of detections in line with frameworks and best practices. You may have even dabbled into custom alerts built from your own internal assessments and findings. Or maybe it’s the opposite; you’re still early in your journey toward security maturity or logging new or custom applications without much guidance. It can be hard to feel truly comfortable with your environment’s security in both situations. Standards are good but can be too noisy and restrictive in some places and too quiet or permissive in others.

In this talk, we’ll explore the cutting-edge work InfusionPoints has done to process and enrich alerts from Graylog using Amazon Bedrock and advanced Large Language Models (LLMs) from Amazon Titan and Anthropic. Discover how we’ve harnessed the power of AI to elevate the accuracy, relevance, and actionable insights of our security alerts, transforming how we respond to potential threats.

The session is really about customers spending their resources wisely, prioritizing use case development based on blind spots, weaknesses, or maybe even just plain audit findings. We have all been guilty in the past of spending a lot of time building clever use cases just for them to never fire or not work out the way we’d hoped; this talk is aimed at highlighting this issue and teach users to focus their resources and build a strategy for development like any other process they would internally.