# Agentic Ai Projects And Mastery
All Agentic Ai Projects And Mastery notes →What is LangGraph
LangChain's graph-based orchestration library — agents as a StateGraph of nodes and edges, with durable checkpointing and human-in-the-loop interrupts as native graph mechanics rather than bolted-on features.
LlamaIndex Workflows
The API surface behind LlamaIndex's event-driven orchestration layer — Workflow/Step/Event class mechanics, Context state and fan-out/fan-in, and how a Workflow sits directly on top of the existing Index/QueryEngine retrieval stack.
OpenAI Agents SDK
Python SDK for OpenAI's Agents framework — the concrete API surface: Agent/Runner construction, the function_tool and handoff() call shapes, Session-backed state, guardrail decorators, and built-in tracing.
AutoGen
Microsoft's open-source framework for conversational multi-agent systems — the concrete API surface: AssistantAgent/UserProxyAgent setup, GroupChat/GroupChatManager wiring, tool registration, and the parameters that actually terminate a run.
Haystack Agents
Haystack's Agent component and Pipeline/Tool wiring — the API surface: how a Pipeline gets built and connected, how Agent slots in as one more Component, and how ComponentTool turns an existing retriever into a tool with no re-integration work.
Semantic Kernel
Microsoft's SDK for adding tool-calling and planning to an existing .NET or Python application — the API surface: Kernel construction, KernelFunction/plugin registration, the automatic-function-calling planner, and where .NET/Python SDK parity actually diverges.