Beyond the SRE: Democratizing observability data with GenAI

Alex Whilhem

Host

The New Stack

Brad Quarry

Senior Principal Solutions Architect

Elastic

Thaddeus Walsh

Principal Solutions Architect

Elastic

Observability data has long been the domain of SREs and IT operations teams. They're the ones who know how to query it, interpret it, and act on it — but that's starting to change. GenAI is making it possible for developers, DevOps engineers, application owners, and business stakeholders to tap into observability data directly, asking questions in plain language and getting real answers without needing to be an expert in the underlying systems.

This shift matters because the questions worth asking aren't only coming from operations. Developers want to debug production systems without waiting on a specialist. Business owners want to know what a 20-minute outage actually cost in revenue. Compliance teams need to investigate personally identifiable information (PII) exposure across historical and real-time data. When observability data is locked behind specialized tooling and knowledge, those questions go unanswered or take far too long to resolve.

In this session, Elastic's Thaddeus Walsh, principal solutions architect, and Brad Quarry, senior principal solutions architect, will walk through how OpenTelemetry brings your observability data together in context and how GenAI unlocks that data for a much wider range of roles and use cases than traditional observability tools allow. You'll see real examples across developers, DevOps, application owners, and compliance teams and leave with a clearer picture of what the right observability platform needs to look like as agentic AI continues to evolve.

Highlights

• Understand how GenAI is expanding, who can use observability data, and what they can do with it.
• Explore use cases across developer debugging, deployment analysis, business impact investigations, and compliance.
• See how OpenTelemetry brings observability data together in a way that makes GenAI-powered querying possible.
• Evaluate what flexibility you need in your observability platform and large language model (LLM) choices to be ready for the future of agentic AI.
• Move beyond reactive incident response toward real-time investigation while systems are running in production.