Join Us at Cisco Live 2026

Assurance for AI: Monitoring OpenAI, Gemini, and More

CISCOU-2071

Speaker:
Abdiel Hernandez Fentanes, Technical Marketing Engineer, Cisco

How do you monitor AI services used by your apps and more importantly, how do you identify a problem in the service delivery chain that could impact the user experience? In this technical session, learn to monitor leading AI platforms—OpenAI, Anthropic, Google Gemini, Azure AI Foundry, AWS Bedrock, and GCP Vertex AI—using ThousandEyes.

See live demonstrations of deploying and visualizing the results and discover how ThousandEyes can provide actionable insights into the performance and reliability of your AI integrations. Whether you’re responsible for AI infrastructure, observability, or service assurance, this session will equip you with practical knowledge to keep your AI-powered applications running smoothly.

Preparing for Agentic Operations: Leadership Strategies for Infrastructure Resilience

ITLGEN-2013

Speaker:
Raj Rikhy, Director, Product Management, Cisco

AI agents are transforming infrastructure management, driving unprecedented speed, automation, and efficiency. As IT leaders expand agent adoption, new opportunities emerge—along with unique operational dynamics. Agents operating at machine speed on shared infrastructure create interaction patterns that didn't exist in human-paced operations. This session explores how organizations can maximize the value of agentic operations while proactively preparing for the challenges of multi-agent, automated systems.

Through analysis of real-world incidents, we’ll reveal how agent-driven environments can amplify interaction failures—where individually correct systems intersect to create unexpected outcomes. The solution lies not in resisting automation, but in evolving infrastructure and leadership practices for this new reality.

From Reactive to Autonomous: Machine-Speed Assurance for AgenticOps

PSOENT-1033

Speaker:
David Puzas, Director, Product Marketing, Cisco

The race to adopt AgenticOps is accelerating, promising networks that can detect, diagnose, and resolve issues in seconds. But automation without trust is dangerous; small mistakes can quickly cascade into outages, compliance risks, and broken digital experiences.

In this session, see how Cisco Assurance enables AI-driven operations to move at machine speed while continuously validating actions across networks, cloud, Internet, and SaaS. Learn how organizations turn AI insights into trusted outcomes, resolve issues faster, and build the machine trust required for autonomous IT.

Assuring the Cloud in the Era of AI

BRKENT-3003

Speaker:
Abdiel Hernandez Fentanes, Technical Marketing Engineer, Cisco

In this breakout, we'll explore how ThousandEyes Cloud Insights delivers end-to-end observability across hybrid network architectures, including Cisco Cloud on Ramp deployments. Understand how traffic actually behaves as it traverses SD-WAN fabrics, cloud on-ramps, and provider networks—and where visibility typically breaks down when things go wrong.

We'll walk through how to correlate ThousandEyes synthetic test data with VPC Flow Logs to distinguish between cloud provider issues, underlay network problems, and misconfigured connectivity policies. See how topology visualization across hybrid links changes how quickly your team can isolate fault domains and make informed decisions under pressure.

Leave this session with a practical framework for monitoring hybrid cloud connectivity, a clearer picture of where ThousandEyes fits alongside Cisco's SD-WAN and Cloud on Ramp solutions, and concrete steps to close the visibility gaps that matter most in your environment.

Failure Patterns in Modern Distributed Systems: Lessons from Major Internet Outages

BRKOPS-2000

Speaker:
Mike Hicks, Principal Solution Analyst, Cisco

As applications evolve into increasingly distributed architectures with specialized services and complex dependency chains, traditional approaches to outage detection and root cause analysis are failing. Through analysis of major incidents across cloud providers, collaboration platforms, and content delivery networks, this session examines three critical failure patterns that explain why outage symptoms increasingly appear disconnected from their root causes.

We'll explore how service features designed for coordination accidentally become failure distribution mechanisms, why systems can appear completely healthy while core functions silently break, and how well-intentioned configuration changes cascade through interconnected systems with devastating effects. Crucially, we'll show how the ThousandEyes MCP server—and the analytical framework it enables—was used to surface, interrogate, and codify these patterns from real incident data, turning raw telemetry into repeatable forensic methodology.

Through detailed case study walkthroughs, attendees will see both the patterns themselves and the investigative process behind them, learning practical strategies to detect and prevent similar failures in their own environments. This 90-minute session is led by the Principal Analyst from ThousandEyes' Internet Intelligence team, who leads outage analysis and publishes findings through The Internet Report. Attendees will leave with actionable frameworks for identifying hidden dependencies, detecting functional failures before users do, and preventing configuration changes from cascading into major incidents.

ThousandEyes MCP in Action: Real-World Use Cases and Integrations to Empower AIOps

BRKOPS-2001

Speaker:
Bernie Clairmont, Technical Marketing Engineer, Cisco

Operationalize the ThousandEyes MCP Server to integrate network intelligence seamlessly into modern AIOps and ITSM workflows. This session presents real-world use cases from MCP client configurations to rapid troubleshooting, enriching incidents and ultimately creating postmortems.

Learn how to leverage conversational AI for natural language queries and build workflows that accelerate mean time to resolution. Empower your network and application teams to collaborate efficiently and resolve issues with precision and speed.