# Eugo for ML engineers

Train and serve models across GPU compute nodes, and understand what the runtime is doing on your behalf.

Source: https://university.eugo.io/learning-paths/ml-engineer

---

## Who it's for

You build and ship models, and your bottleneck is training throughput or inference cost. This path focuses on GPU acceleration, resource requests, and the automatic optimizations that run whether or not you ask for them. Knowing what they do is what lets you work with them rather than against them.

## Courses, in order

1. [Getting started with Eugo](https://university.eugo.io/courses/getting-started-with-eugo) — What Eugo is, what problem it solves, and how to run your first workload on a cluster.
2. [Distributed Python with eugo.hpc](https://university.eugo.io/courses/distributed-python-with-eugo-hpc) — Turn ordinary Python into distributed work with a decorator, then scale it across a cluster.
3. [GPU acceleration on Eugo](https://university.eugo.io/courses/gpu-acceleration-on-eugo) — How work reaches the GPU, which operations benefit, and how to confirm it actually happened.
4. [How automatic optimizations work](https://university.eugo.io/courses/how-automatic-optimizations-work) — Eugo rewrites and tunes your code as it runs. Here is what it does, and how to help it.

## Good to know

### Supported frameworks

Eugo does not support TensorFlow, spaCy, or JAX. Plan around PyTorch and the scientific Python stack.

### GPU availability

GPU compute nodes depend on your plan. Check the dashboard before designing a workload that assumes them.