Quick Notes

miscellaneous concepts that might be helpful in the future :)

Journey learning:

• Includes incorrect solution paths, allowing the model to learn from mistakes.

• From Understanding Reasoning LLMs

Automated Design of Agentic Systems

• Automatical design new agents using a Meta Agent (Foundation Model) to search in the code space (program) and evaluate the result

• Similar discussion seen here: System Prompt Design

• "The history of machine learning teaches us that hand-designed solutions are eventually replaced by learned solutions."