Alexander Zehetmaier

Jonas Vander

OFF CLOUD by Sovereign AI Lab

Xinity AI

A hands-on migration and tuning workshop for DACH engineers moving from cloud APIs to local LLMs.  Date: 28.04.2026, 17:30 – 22:00 PM CET

Mariahilfer Str. 36

<h3>ABOUT THIS SESSION</h3> Your cloud API invoice is climbing. Your compliance team is asking harder questions. You've heard "just run it locally," but nobody has shown you how. And many engineers who already run locally are leaving meaningful headroom on their hardware. This session fixes both problems in the same room. Two tracks: Track A, Cloud to Local. For teams currently using hosted LLM APIs who have never run local inference. You walk in with an API key. You walk out with a working local model, a drop-in API swap, and a cost comparison template you can fill with your own workload numbers. Track B, GPU Whispering. For engineers already running local inference engines who want to go deeper than default settings. Quantization, VRAM layout, CPU memory bus saturation, real benchmarks on real hardware. Cross over during open lab. Ask the awkward questions. <h2>WHO THIS IS FOR</h2><ul><li>Technical founders, CTOs, and engineers whose pay-per-token invoices have become a fast-growing expense line.</li><li>ML and backend engineers who want to get off the API treadmill.</li><li>DevOps leads and infrastructure architects already running local models who want to go deeper than default settings.</li><li>Curious minds exploring sovereign AI</li><li>Sectors: healthcare, legal, manufacturing, financial services, media, startups.</li></ul> <h2>GROUND RULES</h2><ul><li>No sales pitches. Ever.</li><li>Bring a laptop. If you're already running local models, bring that setup. If not, we'll get you started during doors (17:30–18:00), and a pre-event setup gist goes out the day before.</li><li>All skill levels welcome, provided you write code.</li><li>English is the working language.</li></ul>

Agenda

Doors open

Welcome words

Cloud to Local. For teams using hosted APIs who have never run local inference. Install, first inference, port a real API call, measure latency and cost against your current setup. 90 minutes, no prior setup required.

Track A

For engineers ready to go deeper. Quantization, VRAM layout, CPU memory bus saturation, real benchmarks on real hardware. Bring your setup.

OFF CLOUD by Sovereign AI Lab

More events from Xinity AI

Discover more events from Xinity AI, from Science & Tech to other experiences you might love.

Still looking for the right event?

Explore all events in Wien and filter by date, category, and more to find the perfect fit.

OFF CLOUD by Sovereign AI Lab

More events from Xinity AI

Discover more events from Xinity AI, from Science & Tech to other experiences you might love.

You might also like...

Browse more events with different dates, prices, and formats to find your next great experience.

Still looking for the right event?

Explore all events in Wien and filter by date, category, and more to find the perfect fit.