Cloud Native AI Agents in Production: From SRE to Multi-Agent Systems

RSVP Here - https://ocgroups.dev/cncf/group/jqbyur2/event/gkkgzva

Join us for a CNCF Tel Aviv meetup exploring how AI agents are moving from hype to production.
We’ll hear practical lessons from building reliable SRE agents, evolving AI products into multi-agent systems, and navigating the fast-changing architecture of cloud native AI.

Building and Evaluating SRE Agents - Architectural Decisions and Tradeoffs
18:00-18:30 Natan Yellin, Founder/CEO HolmesGPT, Robusta.dev

Building an SRE agent that works reliably in production looks deceptively simple - wire an LLM to MCPs, ship it. The reality is harder.How do you implement memory and learning? What's the cost per investigation, and can you apply the agent to all alerts or only P1s? How do you measure agent performance? Etc.
This talk walks through the architectural decisions we faced when building HolmesGPT, the CNCF SRE agent now running production for everyone from fast growing startups to Fortune 500 customers.
We'll cover topics like:
* Agentic search vs. RAG
* Tool design:
* Context and memory
* Cost / Token budgets
* Evals
* Permissions and multi-tenancy
Whether you're adopting HolmesGPT or building your own SRE agent, we'll share practical tips to make this work in your own production environment

From Prompt to Multi-Agent System: The evolution of our AI product
18:30-19:00 Hila Fox, AI Product Manager, Qodo

Once we were an agent, then an agentic agent, and now a full-blown multi-agent system. When I say 'once,' I mean one year ago. Remember when you could build products that didn't change under your feet every 5 minutes? I do. It was nice.
In this talk, I'll share Qodo's journey through AI's chaos: 'agentifying' our code generation agent, architecting a multi-agent code review system using LangGraph and MCPs, and how we prepare for the inevitable next shift—because in AI, the only constant is change.