Jemiah Sius
New Relic | Senior Director, Developer Relations
Miami, Florida, United States
Actions
I am a Full-stack Engineer turned DevRel Engineer at New Relic. With a background in software engineering and marketing direction, my experience involves: designing and developing user interfaces, addressing customer/client pain points, marketing automation, and community development.
Area of Expertise
Topics
Debugging in the Dark: Building Open-Source Observability for Cloud-Native AI Agents
Traditional cloud-native monitoring wasn't built for autonomous AI agents. A pod might show perfectly healthy CPU and memory while the agent inside quietly burns $5,000 over a weekend looping through failed API calls.
As systems evolve from simple prompts into complex Agent-to-Agent (A2A) and Model Context Protocol (MCP) workflows, standard metrics leave platform teams debugging in the dark. Real-world failures demand a shift toward observing AI decision paths rather than just system health.
This technical deep-dive explores how to implement OpenTelemetry GenAI semantic conventions to solve this blind spot. Attendees will learn to instrument agents across MCP boundaries, manage context propagation for multi-agent handoffs, and achieve step-level token and cost attribution per tenant. Ultimately, we will demonstrate how treating evaluation signals as telemetry prevents runaway costs and builds trust in scalable autonomous systems.
Mastering Problem Solving in Engineering with Observability
Are you an engineer tasked with developing and tackling issues in modern applications? Discover how a strategic approach to problem-solving can enhance your efficiency. This session draws on insights from New Relic’s observability teams and their experiences with distributed systems, providing a structured blueprint for excelling in problem resolution. Learn about the critical tools and methodologies that have proven effective in diagnosing and resolving complex problems in production environments, enabling you to solve challenges more swiftly, effectively, and intelligently.
How to cure generative AI hallucinations with observability
Generative AI (GenAI) applications are known to “hallucinate,” or confidently respond with false or made-up information. Hallucinations are one of the biggest drawbacks to using this tech and are also famously hard to detect, prevent, and fix. But, researchers and engineers are racing to solve this problem as GenAI adoption increases.
So, for a DevOps team adopting this tech now, what’s the best way to tell when hallucinations happen, and most importantly, why?
Engineers, and the company’s bottom line, need a way to trace the path of the response—from the user’s prompt and integration with a large language model (LLM) to work by agents, API calls, and back to the end user. By monitoring every step, engineers can learn where hallucinations originate and apply a potential fix, like adding more context to the prompt or ensuring the LLM is re-summarizing responses accurately, so they can quickly take action and improve the quality of their results.
In this session, Jemiah Sius, Senior Director of Developer Relations at New Relic will explain how companies can use observability to help prevent GenAI hallucinations and share ways DevOps teams can improve their LLM integrations with context, agents, chained models, and more—with a focus on how to do it affordably.
In his role, Jemiah leads a global team focused on solving friction points for engineering teams—keeping them at the center of New Relic’s innovation strategy—so they can more easily find and fix issues before they impact their business and customers.
With that, he understands the challenges DevOps teams face and how, if deployed strategically, GenAI solutions can help solve those challenges.
Seeing the Unseen: How AI Observability Is Redefining Control in the Autonomous Enterprise
This session breaks down what true AI observability demands in high-stakes enterprise environments. We will explore how New Relic delivers deep, real-time insights into distributed AI pipelines, bridging the gap between model behavior, data flows, and infrastructure health. Attendees will learn actionable strategies to detect anomalies before they impact end users, along with a practical framework for implementing AI observability to build trust in autonomous decision-making systems.
- How modern enterprises are evolving toward autonomy, and why traditional monitoring tools can’t keep up.
- What “AI observability” really means in the context of mission-critical operations.
- How New Relic enables deep, real-time insights across distributed, AI-powered environments.
- Detecting anomalies, system state drift, and performance degradation before they impact customers.
- Practical steps to implement AI observability and build confidence in autonomous decision-making systems
Your Pod Is Healthy, Your Agent Is Not: Tracing Agent Behavior End-to-End With OpenTelemetry
A pod can look perfectly healthy on every dashboard while the agent inside it loops through [4,000] tool calls overnight, hands a task to a second agent that silently drops the context, and burns [$X] in tokens before anyone notices. These are behavioral failures, not infrastructure failures, and the request/response trace model most teams run today was never designed to capture them.
This session shows how to instrument a multi-agent system on Kubernetes with OpenTelemetry so that decisions, not just requests, become observable. Using an open-source agent framework, the OpenTelemetry Collector, and an open-source backend, we walk through the GenAI semantic conventions as they stand today, then go past them: propagating context across MCP tool calls and agent-to-agent handoffs, modeling loops and retries as span trees, attributing tokens and cost to the agent step and Kubernetes namespace that caused them, and correlating agent spans with inference-gateway and GPU signals to separate “the model is slow” from “the agent is thrashing.”
Attendees will leave with a reference instrumentation pattern, Collector pipeline configuration for aggregating and sampling agent telemetry, and a clear picture of where the conventions still fall short and how to feed those gaps back to the project.
DeveloperWeek 2027 Sessionize Event Upcoming
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top