Model Customisation
Adapt open models to your task, prove the gain, and serve them efficiently.
Transformers first, so LoRA and the KV cache are mechanisms rather than magic. Then fine-tuning itself, and LoRA-trained models go through LLMOps-style evaluation and release like any other change.
3 courses · 14 lessons · 2h 2m
- Open
Machine Learning & Deep Learning Foundations
Understand the mathematical foundations of Neural Networks, Transformers, Attention, and Fine-Tuning.
4 lessons · Advanced
- Open
Fine-Tuning Open LLMs: LoRA, QLoRA and Deployment
Decide when to fine-tune, build clean training data, train efficiently with LoRA and QLoRA, evaluate for forgetting, and serve the result.
6 lessons · Advanced
- Open
LLM Evaluation & Production LLMOps
Measure, monitor and safely ship language-model features with evaluation sets, calibrated graders, tracing and release gates.
4 lessons · Intermediate
Nothing here is locked. If you already know the earlier material, start wherever you like — browse everything.