Cover Image for Building Agent Memory and Uncovering Bottlenecks with Elastic and Gigamon

Presented by

Meetups for people working with search. Technical talks about full-text search, vector databases, LLMs, ML, observability, and data at scale. Sponsored by Elastic.

Hosted By

1 Going

Building Agent Memory and Uncovering Bottlenecks with Elastic and Gigamon

Name: Building Agent Memory and Uncovering Bottlenecks with Elastic and Gigamon
Start: 2026-09-29T17:30:00.000-04:00
End: 2026-09-29T19:30:00.000-04:00
Location: The Foundry

You Know, for Search

The Foundry

Cambridge, MA

Welcome! To join the event, please register below.

About Event

Join the Boston Elastic User Group and Gigamon on Tuesday, September 29 at The Foundry for an evening focused on AI agent memory and faster root-cause analysis. Learn how to build long-term memory for AI agents with Elastic and how Elastic and Gigamon can help uncover bottlenecks across complex application environments.

Connect with fellow engineers and practitioners while gaining practical insights into AI, observability, and troubleshooting. Pizza and refreshments will be provided. 🍕

⚡️ Interested in giving a talk? ⚡️
Are you interested presenting on your Elastic use case? We welcome 5-10 minute lightning talks, 45-minute deep dives, and everything in between. If you're interested, please send an email to meetups@elastic.co

📅 Date and Time:
Tuesday, September 29th, from 5:30-7:30 pm EDT

📍Location:
The Foundary - 101 Rogers St, Cambridge, MA 02142
We will be in Shop 6

📝Agenda:

5:30 pm: Doors open; say hi and eat some food.
6:00 pm: Building long-term agent memory in Elastic - Susan Chang - Principal Data Scientist at Elastic
6:30 pm: Hiding in Allowed Traffic: Network-Derived Evidence for Modern Threat Detection- Roman Roytfeld - Senior Sales Engineer at Gigamon
7:30 pm: Event ends

💭 Talk Abstracts:
Hiding in Allowed Traffic: Network-Derived Evidence for Modern Threat Detection - Roman Roytfeld - Senior Sales Engineer at Gigamon
Modern applications fail across complex service chains, but observability data alone may not reveal where latency begins. In this technical session, Gigamon and Elastic show how network-derived intelligence and application metadata can enrich Elastic Observability. Learn to map client-to-app dependencies, separate network, application, HTTP, and DNS latency, and isolate degraded service edges across hybrid cloud and Kubernetes. Leave with a practical framework for faster, evidence-based root-cause analysis and clearer ownership across network, platform, SRE, and application teams.

Building long-term agent memory in Elastic - Susan Chang - Principal Data Scientist at Elastic
Details coming soon!

Location