Build Your Own Reasoning Model: A Full Video Course
What it is
This is an educational video course by Sebastian Raschka consisting of eight videos and over 11 hours of content. It guides users through implementing reasoning model techniques, such as evaluation, inference-time scaling, reinforcement learning, and distillation, from scratch using Python and PyTorch on a pre-trained Qwen3 model.
The problem it solves
Lack of accessible, from-scratch educational resources that explain the internal mechanics of reasoning models, specifically how to implement evaluation, RL, and distillation in PyTorch.
What you could use it for
- Learning the fundamental mechanics of how reasoning models are trained and improved
- Implementing reinforcement learning with verifiable rewards in PyTorch
- Understanding the relationship between reasoning models and agent harnesses
Why it matters
Provides a deep, code-level understanding of modern LLM reasoning capabilities, enabling developers to build and customize reasoning models rather than relying solely on black-box APIs.
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.