LangChain
LangChain is an open-source software framework for building applications that compose language models with prompts, data sources, tools, and other components.
Build AI-Enhanced Web Apps
LangChain is a software framework, including its JavaScript implementation, for composing applications that use language models, tools, retrieval, and multi-step chains.
- How it is used
- Orchestrates context-aware language-model workflows behind JavaScript web applications.
- How deep it goes
- Works through modules, composed chains, memory, document pipelines, testing, and integrations without surveying the entire ecosystem.
- Examples
- Uses a streaming weather chain, PDF and DOCX summarization, RAG applications, and fake-model tests.
- What you can do after
- Build it from scratch: compose LangChain.js workflows for chat, summarization, and retrieval, then integrate them with a web UI.
AI Agents and Applications
LangChain is an open-source framework for building applications that combine language models with data sources, tools, and multi-step workflows.
- How it is used
- Assembles reusable model, prompt, data, and retrieval components for summarization engines and document-backed Q&A chatbots.
- How deep it goes
- Develops the object model and progressively richer prompt, summarization, and Q&A implementations; hands adaptive conditional workflows to LangGraph.
- Examples
- Uses a phishing-reminder model call, few-shot number classification, web research summarization, and a document-backed Q&A chatbot.
- What you can do after
- Build it from scratch with LangChain components to create summarization chains and a search-enabled chatbot connected to private documents.
Natural Language Processing in Action, Second Edition
LangChain is a software framework for developing applications that combine language models with tools, data sources, and orchestration logic.
- How it is used
- Connects an LLM to chatbot prompts and conversation logic within an existing NLP application.
- How deep it goes
- Works through model configuration, prompt chains, conversation memory, and observed replies; does not establish a production-ready deployment procedure.
- Examples
- Uses a parent-permission onboarding chatbot and a number-sequence quiz whose feedback contradicts itself.
- What you can do after
- Build it from scratch with LangChain to prototype a prompted, history-aware chatbot and inspect its unreliable replies.
Read how Natural Language Processing in Action, Second Edition explains LangChain →
Data Analysis with LLMs
LangChain is an open-source software framework for developing applications that integrate language models with prompts, data, and external actions.
- How it is used
- Connects language models, prompts, processing steps, and tools for data-analysis applications that need more than one model call.
- How deep it goes
- Works through a classification chain and an agent setup, but defers detailed coverage of the framework.
- Examples
- Uses CSV review sentiment classification and an agent that combines an SQLite sales database with web search.
- What you can do after
- Apply it via LangChain to classify reviews or configure an agent with SQL and web-search tools.
MongoDB 8.0 in Action, Third Edition
LangChain is a software framework for building applications that compose large language models with prompts, tools, memory, and retrieval components.
- How it is used
- Coordinates LLMs, embeddings, external data, and vector search for a RAG chatbot.
- How deep it goes
- Names framework packages and walks through template setup and PDF ingestion, without detailing LangChain internals.
- Examples
- Uses a `rag-mongo` application template and a MongoDB documentation PDF to populate a vector-search collection.
- What you can do after
- Apply it via the LangChain CLI and `rag-mongo` template to ingest PDF embeddings and expose a chatbot chain.
Read how MongoDB 8.0 in Action, Third Edition explains LangChain →
Generative AI in Action
LangChain is a software framework that provides modular components for building applications around large language models and related natural-language services.
- How it is used
- Builds a generative-AI question-answering application over PDFs.
- How deep it goes
- Shows a Python indexing workflow with chunk sizes and overlap, but does not detail production deployment.
- Examples
- Uses a PDF question-answering listing that builds a FAISS index with OpenAI embeddings.
- What you can do after
- Apply it via LangChain to construct a PDF index for question answering.
Learn Generative AI with PyTorch
A Python library for building applications with large language models through tools, abstractions, and component chaining.
- How it is used
- Builds LLM applications that automate workflows and route queries across models, APIs, and task-specific tools.
- How deep it goes
- Explains abstractions, tool selection, and incremental agent construction, then demonstrates code and image tools without detailing library internals.
- Examples
- Demonstrates factual lookup, scientific computation, sine-and-cosine code generation, text summarization, sentiment classification, and image generation.
- What you can do after
- Build it from scratch by combining an LLM, prompt templates, APIs, and custom tools into a zero-shot agent.
Read how Learn Generative AI with PyTorch explains LangChain →
The Art of AI Product Development
LangChain is a software framework for developing applications that compose language models with prompts, retrieved context, memory, tools, and other components.
- How it is used
- Organizes language-model prompts, context, memory, tools, and workflows into application-ready components.
- How deep it goes
- Explains modular prompt development and framework capabilities, while stopping short of detailed implementation code or internals.
- Examples
- Uses Alex's release workflow, reusable prompts, version control, rollback, and agent memory as concrete examples.
- What you can do after
- Apply it via LangChain to structure prompt iteration and connect language models with context, memory, tools, and workflows.
Read how The Art of AI Product Development explains LangChain →
LLMs in Production
LangChain is a software framework for building applications that combine language models with prompts, tools, data sources, and workflow logic.
- How it is used
- Builds language-model pipelines that combine prompts, models, parsers, retrieval, tools, APIs, and workflow logic.
- How deep it goes
- Demonstrates basic chain composition and broader agents, while exposing practical failures without detailing framework internals.
- Examples
- Uses prompt-model-parser chains, web and YouTube search, Python execution, CSV analysis, conversation memory, and local Mistral agents.
- What you can do after
- Builds chains and tool-using agents, hosts them through LangServe, and traces invocations with LangSmith.
Starting Data Analytics with Generative AI and Python
LangChain is a software framework for developing applications that compose language models with prompts, tools, data sources, and conversation state.
- How it is used
- Simplifies API access and local-model integration for generative-AI data-analysis workflows.
- How deep it goes
- Demonstrates installation, model invocation, and memory limitations, but defers detailed configuration to current library instructions.
- Examples
- Uses Python API integration, a local quantized Llama model, installation commands, and a LlamaCpp prompt invocation.
- What you can do after
- Apply it via Python to connect language models, send prompts, capture responses, and add conversation memory.
Read how Starting Data Analytics with Generative AI and Python explains LangChain →
Software Testing with Generative AI
LangChain is a software library and toolset for developing applications that integrate with language models.
- How it is used
- Builds retrieval-augmented testing assistants that supply relevant user stories to language-model prompts.
- How deep it goes
- Demonstrates document loading, query input, cosine-distance retrieval, prompt construction, and model invocation, while noting retrieval limitations.
- Examples
- Uses a Java command-line assistant that retrieves booking user stories and recommends testing risks.
- What you can do after
- Build it from scratch as a Java RAG workflow that loads documents, retrieves context, and sends prompts to OpenAI.
Read how Software Testing with Generative AI explains LangChain →
Machine Learning Platform Engineering
LangChain is a software framework for building applications that compose language models with prompts, tools, retrieval, and other processing components.
- How it is used
- Builds LLM applications by composing retrieval, prompts, models, tools, and processing components into maintainable pipelines.
- How deep it goes
- Explains component interfaces and RAG composition, then stops short of the branching and iterative workflows assigned to LangGraph.
- Examples
- Demonstrates manual RAG, RetrievalQA, FAISS, OpenAI embeddings, Google embeddings, and Langfuse callbacks.
- What you can do after
- Build it from scratch as a composable RAG pipeline with retrievers, chains, validation, source documents, and operational tracing.
Read how Machine Learning Platform Engineering explains LangChain →
Effective Conversational AI
LangChain is a software library for developing applications that compose language models with prompts, data sources, retrieval, and other processing components.
- How it is used
- Simplifies retrieval-augmented generation ingestion by connecting document loading, transformation, and retrieval.
- How deep it goes
- Defines ingestion components and formats, then introduces chunking and search choices without detailing a complete application.
- Examples
- Uses text files, CSV files, directories, PDFs, HTML, JSON, Markdown, web pages, transcripts, and corporate documents.
- What you can do after
- Operate and configure an ingestion pipeline that loads documents and retrieves relevant chunks for language-model responses.
Data Storytelling with Altair and AI
LangChain is a software framework for building applications that connect language models with prompts, external data sources, and other components.
- How it is used
- Connects language models to prompts, indexes, and external data for context-aware applications.
- How deep it goes
- Explains chains and components, then demonstrates loading, chunking, embedding, retrieval, prompting, and querying.
- Examples
- Uses a smartphone HTML page, Chroma collection, OpenAI embeddings, a constrained prompt, and a RetrievalQA query.
- What you can do after
- Build a retrieval-augmented application that answers product questions from indexed HTML context.
Read how Data Storytelling with Altair and AI explains LangChain →
The Complete Obsolete Guide to Generative AI
LangChain is a software framework for building applications that coordinate large language models with tools, data sources, and multistep operations.
- How it is used
- Coordinates LLMs with search, math, and document-retrieval tools for multilayered research and analysis.
- How deep it goes
- Demonstrates runnable agent and retrieval pipelines but does not explain their underlying model or vector-store internals.
- Examples
- Uses web research about David Clinton’s publications and compares Alphabet, Cisco, and IBM financial filings.
- What you can do after
- Apply it via LangChain to build Python agents that search the web or compare multiple PDF documents.
Read how The Complete Obsolete Guide to Generative AI explains LangChain →
Build Applications with Local AI Models on a Mac
LangChain is a software framework for building applications with language models by connecting them to data sources and composable components.
- How it is used
- Supports chatbots that need retrieval, memory, tools, or multistep workflows using reusable language-model components.
- How deep it goes
- Introduces components and demonstrates one local model invocation, but defers a full LangChain treatment.
- Examples
- Shows a Python invocation of a local Ollama model with a capital-city question and suggests a document question-answering project.
- What you can do after
- Apply it via LangChain and Ollama to connect a local model, then extend applications with chains, agents, or tools.
Read how Build Applications with Local AI Models on a Mac explains LangChain →
Just Use Postgres!
LangChain is a software framework for building applications that compose language models with prompts, tools, data sources, and processing steps.
- How it is used
- Connects Python applications, prompts, retrieved Postgres context, and language models for generative-AI question answering.
- How deep it goes
- Demonstrates prompt construction, model invocation, RAG flow, and stateless-history handling without deeper framework internals.
- Examples
- Uses TinyLlama, an OllamaLLM interface, a movie prompt, and retrieved Pirates of the Caribbean records.
- What you can do after
- Apply it via Python to build a RAG question-answering function backed by Postgres and an Ollama-served language model.
Investing for Programmers
LangChain is a software framework for building applications that compose large language models with prompts, tools, data sources, and workflow logic.
- How it is used
- Builds Python workflows that combine LLM prompts with provider integrations, external data, and agent tools.
- How deep it goes
- Explains provider abstraction, invoke(), one-shot prompts, and tool-using agents through executable listings.
- Examples
- Demonstrates a Google SWOT prompt and a Yahoo Finance News agent answering what happened with Microsoft stock.
- What you can do after
- Apply it via Python to invoke multiple LLM providers and build agents that retrieve financial information through tools.
Building Reliable AI Systems
LangChain is an open-source software framework for building applications that coordinate language models with prompts, data sources, tools, and retrieval workflows.
- How it is used
- Connects language models, prompts, data, and tools to prototype RAG applications with less integration work.
- How deep it goes
- Explains component roles and helper use for RAG, while deferring detailed retrieval-chain and question-answering features.
- Examples
- Uses a simple RAG system, document loading and text splitting, and a chatbot retrieving product information.
- What you can do after
- Apply LangChain to connect components and prototype a RAG workflow or chatbot.
Related concepts
- language modeluses · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
- retrieval-augmented generationuses · Retrieval-augmented generation is a technique that retrieves relevant information from external sources and supplies it as context to a generative model when producing an output.
- contractuses · An API contract is a specification of the operations, data, and behavioral expectations that govern interactions between software components.
- prompt templateuses · A prompt template is a reusable structure for organizing instructions, context, and variable inputs supplied to a language model.
- LangGraphused by · LangGraph is a framework for building stateful, graph-structured workflows and applications, including those involving language models and agents.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.