Building With LLMs

The course about everything around the model: data, testing, deployment, monitoring, and cost.

Speech recognition, audio classification and text-to-speech with open models.

Vision transformers, generative models and multimodal work, taught with runnable notebooks.

The reference course for the Transformers library: tokenisers, fine-tuning, and shipping a model.

A structured path from what an embedding is through to deploying a working retrieval system.

An interactive course where you edit the code inside the video player. Good if reading documentation bores you.

Production ML taught as software engineering — testing, CI, reproducibility, not just notebooks.

Dozens of free one-to-two hour courses built with the companies making the tools. The fastest way to learn one specific thing.

Free practitioner courses on training, evaluating and instrumenting models in production.

A long-form certificate covering deep learning frameworks and deploying models as services.

AWS's training library, including free AI practitioner paths and Bedrock-specific courses.

NVIDIA's hands-on workshops with GPU environments provided, plus certification.

Google Cloud's official generative AI path, with hands-on labs against real infrastructure.