Module 09
Deep learning and LLMs
Neural network fundamentals, transfer learning, transformers, and building retrieval-augmented applications on top of large language models.
Outcome
What you will be able to do
You can fine-tune a pretrained model and ship a grounded LLM feature with retrieval and evaluation.
Lessons
Work through these in order
- 01Neural network fundamentalsLayers, activations, loss, backpropagation and the training loop — what every framework is doing under the abstraction. 30 min
- 02Transfer learning that actually worksPretrained backbones, freezing versus fine-tuning, augmentation, and parameter-efficient fine-tuning with LoRA. 26 min
- 03Transformers and embeddingsAttention, tokenisation, context windows, and using embeddings for search, clustering and deduplication. 28 min
- 04Building LLM applicationsPrompting, structured output, retrieval-augmented generation, tool use and how to evaluate a non-deterministic system. 32 min
Assessment
Module quiz — 70% to pass
8 questions mixing concept checks and short code-output problems. Graded on the server with per-question explanations afterwards, unlimited retakes, and a badge with a verification code the moment you pass.