Yaolin Ge

Teaching · knowledge sharing

Machine learning, from first principles to production

A curriculum of about fifty lessons written for colleagues, from autograd built by hand to GPT, LoRA, RAG and MLOps, alongside a Python course I wrote and taught.

numpy & autograd supervised unsupervised reinforcement genai mlops lesson 1 lesson 50

Why

A team moves at the speed of its shared understanding. Teaching is also the fastest way I know to find the gaps in my own.

What it covers

  1. Foundations: NumPy, and automatic differentiation written from scratch.
  2. Supervised and unsupervised learning.
  3. Reinforcement learning.
  4. Generative AI: byte-pair encoding, attention, a small GPT, LoRA fine-tuning, retrieval-augmented generation and agents.
  5. MLOps: MLflow, FastAPI serving, Docker, Kubernetes, and Prometheus and Grafana monitoring.

Before this I wrote and taught PyPioneers, a Python course for colleagues. I also led my section's adoption of LLM-based developer tools.