Vibe Coding an LLM Agent All the Way to Production
Join us for an engineer-first evening diving deep into the practicalities of building and deploying LLM agents. Whether you're just starting out or scaling GenAI infrastructure, this meetup is designed to provide actionable insights from industry experts.
Speakers:
🎤 Vibe coding an AI Assistant
In this talk Jonathan will demonstrate how to very quickly and efficiently leverage vibe coding to build an AI assistant on top of Github copilot.
👨💻 Jonathan Yarkoni, CEO of LATENT AI
Jonathan comes from a background in software engineering having previously been an engineer and team lead. He spent 6 years at Google as an SE AI/ML specialist. In the past 2 years Jonathan has been leading LATENT AI a Generative AI professional services company.
🎤 Monitoring and Evaluating an AI Agent
So you’ve built your AI Agent - now what? We’ll explore the different techniques to make sure your agent works as intended. How to measure its performance, improve it over time, and upgrade to newer models.
👨💻 @Gal Kleinman, co-founder and CTO of Traceloop
Traceloop is a platform for observability and evaluation in generative AI. Before starting Traceloop, Gal led platform engineering at Fiverr, where he built and scaled their ML and data infrastructure. Today, he’s focused on helping teams ship reliable LLM-powered apps to production.
🎤 LLM Agents in the Wild: Architectures, Threats, and What Actually Works
LLM agents break in unexpected ways when they hit the real world. This talk covers the architectures, threats, and lessons we’ve learned from taking agents to production. If you're building GenAI systems, you’ll leave with practical insights you can apply right away.
👨💻 @Lior Ziv, Co-founder and CTO of Lasso
Lasso is a startup securing GenAI use in workplaces and production by protecting interactions with LLMs. After 6 years of military service in an elite Israeli Air Force technological unit, Lior led software roles at SAP and cybersecurity startups. He brings deep experience in building scalable, enterprise-grade products across cloud and on-premise environments, covering the full product lifecycle from design to deployment.
🎤 How AI builders can use AWS Strands to build and run AI agents with Nova on Bedrock
The AWS Strands Agents SDK simplifies building AI agents with models like Nova on Bedrock, making development faster and deployment easier.
👨💻 Moshe Shelly, Principal GenAI/ML Solutions Architect, AWS
Moshe is a seasoned software engineer and tech leader with over 20 years of experience in cloud architecture, distributed systems, and security. At AWS, he has served in key roles including Head of Technology for the Enterprise segment and now leads as a Principal Solutions Architect specializing in GenAI and ML.
This is not a panel. Not a pitch. Just builders sharing the real stuff.
If you're coding, shipping, or scaling GenAI systems, you’ll want to be in the room.