Multi Agent Frameworks (LLM)

atomic-agents (OSS, simple, based on instructor

- instructor and schema names = Bad Schemas could break your LLM Structured Outputs - Instructor

magnetic-one (Microsoft)

TinyTroupe  (Microsoft)

TinyTroupe is an experimental Python library designed to simulate individuals, referred to as TinyPersons, with distinct personalities, interests, and goals.

amazon - https://github.com/awslabs/multi-agent-orchestrator

    - https://aws.amazon.com/bedrock/agents/

pydantic fw - https://ai.pydantic.dev/

(related) Gemini - has Plans and Approvals? = https://developers.googleblog.com/en/the-next-chapter-of-the-gemini-era-for-developers/

Scaling Functionality (large number of FCs) = github.com/microsoft/semantic-kernel/discussions/9947
- use a Vector store
- semantic search to find the related function calls, so only send the related ones to the LLM

Semantic Kernel - Agents = https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/?pivots=programming-language-csharp#what-is-an-ai-agent

AI Engineering[Book] [recommended by Zeki]

Anthropic - Building Effective Agents = https://www.anthropic.com/research/building-effective-agents

Agent paper by Google = 1735887787265