Hands-On AI Build a Generative Language Model from Scratch

This lab-oriented course walks through building a small generative language model, covering data collection, tokenization, transformer implementations, training loops, and basic inference optimisation. You will experiment with training pipelines, evaluate outputs using standard metrics, and implement simple optimisations for inference speed and memory. Emphasis is on understanding internals to make informed architecture and training choices.

• Skills you’ll learn: transformer implementation, training pipelines, tokenization, evaluation metrics, inference optimisation.

Certificate: Certificate of completion provided upon finishing the course.

Requirements

• Python and ML libraries (PyTorch or TensorFlow); GPU resources recommended.

Who this course is for

• ML students; researchers; engineers.

Benefits

• Implement transformer blocks; train on a dataset; evaluate model outputs; optimize inference performance.