psychology

Agentic AI — Learning Path

From zero to multi-agent systems. Each chapter: high-level overview + under-the-hood deep dive.

Co-Created by Kiran Shirol and Claude

Core Stack LangChain LangGraph Chainlit LangSmith

home Learning Portal play_arrow Start Learning summarize Key Insights dictionary Glossary 9 chapters · Each with High Level + Under the Hood

Foundations

Stack, Chains & Prompts

The building blocks — from tech stack overview to structured inputs and outputs.

layers

Tech Stack Overview

LangChain, LangGraph, Chainlit, LangSmith — what each does and how they connect.

visibility High Level code Deep Dive

link

Your First LLM Chain

Prompt in, response out. The simplest agent call with ChatPromptTemplate and ChatModel.

visibility High Level code Deep Dive

data_object

Prompt Templates & Output Parsers

Structured inputs, Pydantic validation, and schema-enforced LLM responses.

visibility High Level code Deep Dive

Capabilities

Tools, Functions & RAG

Giving agents the ability to act on the world and access external knowledge.

build

Tools & Function Calling

The @tool decorator, ToolMessage protocol, and MCP for standardized tool access.

visibility High Level code Deep Dive

Retrieval-Augmented Generation

Embeddings, vector stores, and retrieval chains — giving the LLM context it wasn’t trained on.

visibility High Level code Deep Dive

Orchestration

Graphs, Multi-Agent & Observability

Stateful flows, multi-agent collaboration, debugging, and the framework landscape.

account_tree

LangGraph: Stateful Agent Flows

State machines, conditional edges, cycles, checkpoints, and human-in-the-loop patterns.

visibility High Level code Deep Dive

groups

Multi-Agent Orchestration

Supervisor patterns, agent handoffs, shared state, and A2A protocol basics.

visibility High Level code Deep Dive

monitoring

Observability & Debugging

LangSmith traces, Langfuse, evaluations, cost tracking, and production debugging.

visibility High Level code Deep Dive

compare

Framework Landscape

LangGraph, CrewAI, AutoGen, Agno, smolagents, Pydantic AI, Google ADK, and more.

visibility High Level code Deep Dive

Agentic AI — Learning Path

Stack, Chains & Prompts

Tools, Functions & RAG

Graphs, Multi-Agent & Observability

Explore Related Courses