Wed, Jun 24 · 6:30 PM EDT
We're back for our June meetup on Wednesday evening, June 24!
The topic this month is one in which there is a ton of interest, and it's one of special interest to our community - Running Agents Locally with Open Source: From Laptop to Kubernetes.
The speaker is Legare Kerrison, Developer Advocate at Red Hat. Legare is fantastic and will deliver a ton of value to attendees.
Doors will open at 6:00 pm, live programming will begin at 6:30 pm sharp. We'll wrap up no later than 8:00 pm as we know it's a work night for everyone.
Location:
Cisco's RTP campus
Building 10
7200-10 Kit Creek Road
Morrisville, NC
FREE attendance
Attendance is FREE, as always.
Free Swag and food and beverages
Swag (shirts, hats, stickers) and food and beverages will be served onsite, but quantities will be limited, so arrive early if interested. *Food and beverages will be available courtesy of the wonderful team at IBM*
Speaker: Legare Kerrison, Developer Advocate, Red Hat
Bio:
Legare Kerrison is an Open Source Engineer and Developer Advocate on Red Hat's AI team. She focuses on open source tools for building and deploying AI. Currently, she works with projects like OpenShift AI and vLLM. She aims to make technical complexity digestible. She loves matcha and the outdoors. Based in Boston.
Talk Title:
Running Agents Locally with Open Source: From Laptop to Kubernetes
Abstract:
AI agents are everywhere, but most developers are building on top of proprietary APIs, paying per token, and hoping their sensitive data doesn't end up in someone else's training set. When things work, they work great. When you hit a rate limit, a price hike, or a data privacy wall, suddenly your production agentic application becomes a liability.
The good news? The open source ecosystem has caught up with the release of open source models like Qwen 3.5. You can now run fully capable agentic AI systems, locally, on your own infrastructure, and at scale on Kubernetes, without compromising on model quality or developer experience.
In this session, we'll build a working AI agent from scratch using an open source stack: Ollama for local inference, Podman to containerize the entire stack, and Model Context Protocol (MCP) to wire up tools your agent can actually use. Once it's working locally, we'll show the path to scaling it up onto OpenShift/Kubernetes, keeping the same architecture that worked on your laptop.
Questions? Contact us at info@allthingsopen.org.
RSVP required
*This is a free event. However, an RSVP is required. Seating is limited, so RSVP's will be capped as we fill up. Please move quickly and RSVP to save your spot if interested. When you RSVP you will be added to our mailing list to receive future event updates. You can unsubscribe at any time.