Video course · 7 chapters · 15 min · certificate
Understanding and Effectively Using AI Reasoning Models
Lance Martin (LangChain) explains reasoning models like OpenAI o1 and o3: the old next-word scaling paradigm, why chain of thought helps, the new paradigm of reinforcement learning on chain of thought, how to prompt reasoning models, API features like reasoning effort, structured outputs and tools, and the best use cases.
What you'll learn
- Explain the next-token scaling paradigm
- Describe system 1 vs system 2 thinking and chain of thought
- Explain RL on chain of thought
- Prompt reasoning models with goals, not methods
- Use reasoning effort, structured outputs and tool calling
- Choose good use cases for reasoning models
Chapters
7 chapters · 15:16-
1:42
01Scaling Members
The old scaling paradigm
Next-word prediction.
-
1:33
02Reasoning Members
Limitations and reasoning
System 1 vs system 2.
-
2:37
03RL Members
The new reasoning model paradigm
Reward correct chains.
-
1:09
04Prompting Members
Prompting reasoning models
Say what, not how.
-
2:32
05API Members
API usage and features
Effort, structure, tools.
-
3:48
06Use cases Members
Use cases and applications
Where they shine.
-
1:55
07Summary Members
Summary and outlook
Different paradigms.
Jobs this course suits
Our AI checked this course against 500 jobs; these get the most out of it. Each job links to its learning path.