DeepYardDeepYard
A

AI Self-Evolving Agent

Agent that improves its own performance through self-optimization loops

Open SourceFree

About

A self-improving AI agent that evaluates its own outputs, identifies weaknesses, and iteratively refines its strategies and prompts. It implements a feedback loop where the agent scores its performance, generates improvements, and applies them in subsequent runs — demonstrating meta-learning capabilities without human intervention. Part of the awesome-llm-apps collection.

Details

Typeautonomous-agent, self-evolving
Deploymentlocal
Supported Modelsgpt-4, claude

Tags

autonomousself-improvingmeta-learningpython