Frankfurt | Build an AI SRE with Grok
The GlassFlow breakout at GrokBot x Build Days. In about 45 minutes you build two agents on Grok that watch a small live service, break the service on purpose, and watch the agents find the root cause. Then you point the same pattern at whatever you are building today.
- When: Friday 30 October 2026, during the build day (10:00 to 18:00). The breakout runs as a loop: join at the start of any round.
- Where: The Squaire, Am Flughafen 12, Frankfurt. Look for the GlassFlow breakout area.
- Event: GrokBot x Build Days @ AI Week, hosted by SpaceXAI Frankfurt, with SpaceXAI and GlassFlow as community sponsors. Event page
- Bring: a laptop and an xAI API key. Your event Grok credits work for this.
Do the setup before you come. The images for the demo are about 600 MB, and the venue Wi-Fi is shared by hundreds of people.
Tares is open source. The breakout runs on it. If you like what you build, star the repo on GitHub: it helps others find it.
★ Star Tares on GitHub
The breakout
- You watch a demo service with steady traffic: its Prometheus metrics, its error logs and the alerts Prometheus fires. Tares puts all of it on one timeline for the service.
- You build a watcher on a small, fast Grok model that looks every 2 minutes, and a root-cause agent on a bigger Grok model that only runs when the watcher says “investigate”.
- You see a finding appear 2 to 4 minutes after you break the service: what is failing, since when, the likely cause and a next step.
- You take home the watch-then-escalate pattern, and how to give the agent you are building today the same timeline over MCP.
Each round
| Minutes | What |
|---|---|
| 0 to 5 | A live incident on our screen: break the service, watch Grok find out why |
| 5 to 10 | Tares in 5 minutes: sources, the timeline, watch then escalate |
| 10 to 35 | You build it on your laptop, with us walking the room |
| 35 to 45 | Use it on your build: the same timeline over MCP for your own agent |
Rounds start at about 11:00, 13:00, 14:30 and 16:00. Ask us at the breakout for the next one.
Downloads
| File | What it is |
|---|---|
| tares-demo.zip | The demo stack: docker-compose.yml and inject.sh. Unzip it, then run docker compose pull inside: the images (about 600 MB) still come from the internet. |
Keep going
- Build an AI SRE : the same demo, then how to swap in your own metrics, logs, alerts and deploys.
- Watch, then escalate : the watcher and root-cause pattern, for any sources.
- Connectors : Prometheus, Alertmanager, Docker logs, OpenTelemetry, GitHub, Postgres, webhooks and more.
- Connect your agents to Tares over MCP.
- Tares docs and the Tares repo on GitHub .
- xAI API docs .
Last updated on