Own Your AI: Build Your First Model Factory - Hack Night
Build an ML Engineer Agent that evaluates, fine-tunes, deploys, and improves open models on Union.
We'll start with a hands-on walkthrough where we build a LangGraph-powered agent that evaluates open models in parallel, fine-tunes promising candidates, deploys the winner, and retrains when performance drops.
Think of it as the start of your own model factory: a repeatable system for evaluating, improving, and deploying models you control instead of depending on an external API for every task.
We'll also see how Union gives workflows infrastructure as context, so they can adapt when resources change, recover from failures, and keep moving without restarting completed work.
After the walkthrough, we'll switch into hacknight mode: keep building on the example, try your own models or datasets, experiment with evals and fine-tuning, or start something new around the same ideas.
For teams thinking about sovereign AI, this is a practical starting point for owning more of your model stack: the weights, the training process, the evaluation loop, and the infrastructure it runs on.
Bring a laptop!