AI Agent

AI Agent

Creator
Creator
Seonglae ChoSeonglae Cho
Created
Created
2023 Apr 12 14:14
Editor
Edited
Edited
2025 Feb 13 22:20

Generative Agent

  • memory
  • tool interface

Specifically

  • Tool learning
    • System Calls
      • Socket
    • Headless Browser access to pretend human
    • CLI access
      • Curl
        • GET - Search
        • POST - Communication
  • Permanent Storage like vector database
    • Context Selection based on vector similarity
    • Long-term memory by weight update
    • Vanishing to prevent manage capacity based on LLM decision or dynamic sparsity
  • Self-replication, Self-restart
    • Self-initialization prompt, Self-editing
  • Asynchronous communication port
    • Interrupt

Text to Task
Generative Model

  • Photoshop like AI software
AI Agent Notion
 
 
 
 
https://arxiv.org/pdf/2304.03442
https://github.com/phidatahq/phidata
 
 
 

Efficient agents

While workflows orchestrate LLMs and tools through predefined code paths, Agent LLMs determine tool usage and processes. In other words, Prompt Chaining and Parallelization are workflows, while agents do task planning and execution for unpredictable task paths. Workflows are used when predictability and consistency are needed, while agents are suitable for large-scale tasks that require flexibility and model-driven decision making.

Google whitepaper

Cognitive architecture

Examples

AI Agency

 
 

Recommendations