LLM Engineering Lab: Build with Open Models

the vibe

Friday, November 20, 2026 · 10 AM–6 PM PT GitHub Bellevue · 10900 NE 4th St, Floor 21, Bellevue, WA 98004 A workshop and build day with HackerSquad and Akamai Inference Cloud What changes when you're the one running the model? Join HackerSquad and Akamai Inference Cloud at GitHub for a full day of hands-on LLM engineering. We'll explore how to run open-weight models on GPUs, experiment with the choices that shape their behavior, and put those ideas to work in a project of your own. The morning will focus on guided learning. In the afternoon, you'll have time to test ideas, build, and compare what you discover with other developers. We'll close with demos and a discussion of what worked, what didn't, and what to try next. What we'll explore Running open-weight models: Get a model running on GPU infrastructure and understand the resources it needs. Quantization: Experiment with reducing model precision and examine the effects on memory use and output quality. Context windows: Explore how much context fits alongside your model and what that means for the application you're building. Putting it into practice: Turn an experiment into a small application or demo, then share what you learned. Who this is for Developers, AI engineers, and technical founders who want practical experience working directly with open-weight models. You should be comfortable writing basic Python and working in a terminal; prior experience running your own models isn't required. Bring a laptop, charger,…