# No Evals, No Optimization, No Production.
**Date de l'événement :** 27/05/2026
* Publié le 27/05/2026

### Date
27/05/2026

### Galerie d'image
![1.png](https://firebasestorage.googleapis.com/v0/b/memory-ai.appspot.com/o/prod%2FrKxsdSTpqCfzIFY8Y2hg%2FprojectsMedias%2FhcvvEiLIlauzLljCwsGX%2Fthumbs%2F1_1600x900.png?alt=media&token=13e6035c-d1f3-43ee-83d4-fdca409caa06) 

### Ville
`#San Francisco` 

## Description
AI agents are moving from demos into production. But once they start touching real workflows, teams face a new set of questions:  
Are they doing what we expect?  
Where do they break?  
How do we trace failures back to the right step?  
And once we find the problem, how do we actually improve the agent?  
This meetup is for builders who want to move beyond “it seems to work” toward measurable, debuggable, and continuously improving AI agent systems.  
What We’ll Cover  
How to evaluate AI agents beyond simple pass/fail checks  
Why observability matters once agents move into production  
How runtime traces help teams understand where agent behavior starts to break  
How to connect agent failures back to workflow steps and code context  
How evaluation and debugging signals can become optimization inputs  
How AI agent systems can improve through iterative optimization loop  
What to expect:  
5:30 - 6:00: Arrival, Pizza, Drinks, Networking  
6:00 - 6:20: Talk – Runtime debugging for agents, with step-level reproduction, handed off to your agent via MCP.  
6:20 - 6:30: Demo – Finding the failing traces, understanding what went wrong, and patching the code, all through MCP.  
6:30 - 6:50: Talk - How agent optimization moves teams beyond debugging individual failures  
6:50 - 7:00: Demo - MEGA Workbench turning eval signals into iterative performance gains.  
7:00 - 8:00: Q&A, then Networking  
Pizza and beverages will be provided.  
  
Save your spot, spaces are limited.  
Venue graciously provided by HAC.  
About the Hosts  
MEGA Code  
MEGA Code builds self-evolving AI agent optimization infrastructure, focused on evaluation-driven development, reusable agent wisdom, and optimization loops that help AI agent systems improve across runs.  
Elastic Dash  
ElasticDash is a step-level debugging tool for AI workflows, helping developers identify and resolve issues at each stage of their agent and LLM pipelines.  
HAC  
Hanwha AI Center (HAC) is a private membership for all those dedicated to AI. As a hub for innovation, HAC brings together passionate entrepreneurs, researchers, and visionaries to explore the societal and technological impacts of AI on human life.

**Lien de l'évènement :** [https://luma.com/9kag32ur](https://luma.com/9kag32ur)

### Pays
`#United States` 

### Continent
`#North America` 

**Médias associés :**
[Média 1](https://80954c1d.sibforms.com/serve/MUIFABojU8UBbDiX_TdcGa7Wv5VMoVB_nBZ92mkLkGlS1pJLpP7s-pVJusyN-7cG9KPrSuv3fv7TmXwuw_AoyNUShR8jZhmNDgUbZPJO2V5xYXlNz4YXOTjSb8X7Lj7PRIPzgzEWlLbA4f4uw_F8RM51EUsjSfQQko0qaby98GHMdYJVWLIXd5JzzaXBGmqN2CcYOFuqnbnaYEnw) 

## event_id
evt-t5F9sky0kbSHh2l@events.lu.ma

### Outils
`#MCP` 

### Selection
`#Agents` 



---
### Navigation pour IA
- [Index de tous les contenus](https://ai-memory.io/llms.txt)
- [Plan du site (Sitemap)](https://ai-memory.io/sitemap.xml)
- [Retour à l'accueil](https://ai-memory.io/)