# How to Run Your Own AI: Real-Life Experiences Operating a Self-Hosted Large Language Model for SeaTable Automations

- Institution: CERN
- Speaker: Christoph Dyllick-Brenzinger
- Year: 2026
- Language: English
- Subjects: Computer science
- Source: [CERN](https://videos.cern.ch/record/3025479)
- Licence: [CERN](https://copyright.web.cern)
- Who can watch: Tweens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

Self-hosting a large language model (LLM) to power AI-driven automations is no longer just a theoretical possibility—it's a practical reality. In this talk, discover how SeaTable 6.0 integrates AI-powered automations by running a self-hosted LLM on a GPU server at Hetzner using the vLLM framework. Learn about the technical challenges, infrastructure requirements, and operational insights gained from deploying and managing your own AI model in a production SaaS environment. This session offers a unique, hands-on perspective for organizations looking to leverage AI while maintaining full control over their data and customization. Whether you're exploring AI adoption or seeking alternatives to cloud-only providers, this presentation shares actionable lessons and best practices from real life.

00:00:00 Slide 1

00:00:05 Slide 2

00:01:12 Slide 3

00:01:30 Slide 4

00:02:51 Slide 5

00:04:02 Slide 6

00:04:33 Slide 7

00:06:41 Slide 8

00:07:40 Slide 9

00:08:30 Slide 10

00:10:01 Slide 11

00:10:46 Slide 12

00:11:38 Slide 13

00:12:38 Slide 14

00:14:14 Slide 15

00:15:04 Slide 16

## Lectures

1. [How to Run Your Own AI: Real-Life Experiences Operating a Self-Hosted Large Language Model for SeaTable Automations](https://invisiblenetwork.ai/learn/cern/3025479/3e7ccf358315), 16 min (CERN's player)