Learn LangChain

Learn LangChain v1.x from the fundamentals to production-grade: prerequisites & environment setup, history & background, core concepts & main architecture, model components, prompts & templates, output parsers, chat models, embeddings & vector stores, chains, LCEL (LangChain Expression Language), RAG with LangChain, tools & tool calling, agents, memory & state, observability with LangSmith, deployment with LangServe, LangGraph for agentic workflows, streaming & async, fine-tuning & evaluation, error handling & retry, security & best practices, modern features & roadmap, and alternative ecosystems & final reflections across a total of 23 episodes.

⌘ + /

Sponsor this project

Back to Series

0 views

Read

Learn LangChain - Prerequisite Skills & Environment Setup

The opening episode of the Learn LangChain series: the basic Python 3.10+ skills and LLM concepts (tokens, temperature, prompts, API keys) you must master, followed by environment setup with venv/uv, installing the langchain and langchain-openai packages, and verifying that the setup is ready to use.

August 3, 2026

AI Agent

Read

Learn LangChain - History, Background & Why LangChain Is Needed

From open source in 2022 to v1.0 in 2025: tracing LangChain's evolution from a simple chaining framework into an agent engineering platform, understanding the problems it solves — provider standardization, composability, memory, all the way to stateful agent orchestration.

August 3, 2026

AI Agent

Read

Learn LangChain - Core Concepts & Main Architecture

Breaking down the LangChain v1.x architecture: the core langchain-core primitives (chat models, prompt templates, output parsers, tools, embeddings, vector stores, retrievers), the LCEL composition language with the pipe operator, and the package structure map that is now modular per provider.

August 3, 2026

AI Agent

Read

Learn LangChain - Setup, Chat Models & Invocation

First hands-on practice: full project setup with API keys from .env, initializing ChatOpenAI and ChatAnthropic, then all invocation modes — invoke, ainvoke, stream, astream, batch, abatch — along with the contents of the response object such as content and token usage.

August 3, 2026

AI Agent

Read

Learn LangChain - Advanced Prompts & Chat Models

Going deeper into prompts and chat models: PromptTemplate versus ChatPromptTemplate, system and human messages, few-shot prompting, input variables, per-provider model profiles, and content-block centric v2/v3 streaming with version="v3" in stream_events.

August 3, 2026

AI Agent

Read

Learn LangChain - LCEL: Chains & Composition

Dissecting LCEL thoroughly: RunnableSequence with the pipe operator, RunnableParallel for parallel branches, RunnablePassthrough and RunnableLambda, RunnableMap, the invoke/batch/stream lifecycle with async variants, event streaming, and error handling with retries.

August 3, 2026

AI Agent

Read

Learn LangChain - Output Parsers & Structured Output

Dissecting output parsers in LangChain: StrOutputParser, JsonOutputParser, PydanticOutputParser, and with_structured_output to force the model to return a structured format, complete with streaming parsing techniques for partial output.

August 3, 2026

AI Agent

Read

Learn LangChain - Memory & Conversation State

Building agents that remember: summarizing and storing conversation history, using thread_id for parallel sessions, and the LangGraph Checkpointer for snapshots and resuming conversations across requests.

August 3, 2026

AI Agent

Read

Learn LangChain - Tools & Function Calling

Giving your agent the ability to act: defining tools with the @tool decorator, binding them to a model via bind_tools, and understanding the selection, execution, and error handling loop of tool calling.

August 3, 2026

AI Agent

Read

Learn LangChain - Document Loaders & Text Splitters

Converting various document sources into Document objects and then splitting them into chunks: RecursiveCharacterTextSplitter, token-based splitters, chunk size and overlap, and the right chunking strategy for RAG.

August 3, 2026

AI Agent