The LangChain ecosystem

Developing LLM Applications with LangChain

Jonathan Bennion

AI Engineer & LangChain Contributor

Meet your instructor...

Jonathan Bennion headshot.

Jonathan Bennion, AI Engineer
ML & AI at Facebook, Google, Amazon, Disney, EA
Created Logical Fallacy chain in LangChain
Contributor to DeepEval

Build LLM Apps with LangChain

A building site for LLM/AI applications.

The LangChain ecosystem consisting of LangChain, LangSmith, and LangGraph.

LangChain integrations

LangChain partner organizations, including Anthropic, OpenAI, Hugging Face, neo4j, Databricks, XAI, Mistral, Cohere, Snowflake, and AWS.

¹ https://python.langchain.com/docs/integrations/providers/

Building LLM apps the LangChain way...

A customer support chatbot with two different modes: one to provide product information and make recommendations, and another to provide specific advise and FAQs for customers experiencing issues with placing orders.

Building LLM apps the LangChain way...

An LLM represented as a car engine.

Building LLM apps the LangChain way...

A split decision.

Building LLM apps the LangChain way...

A database.

Building LLM apps the LangChain way...

A magnifying glass representing the finding and retrieval of information.

Prompting OpenAI models

from langchain_openai import ChatOpenAI


llm = ChatOpenAI(
    model="gpt-4o-mini",
    api_key='...'
)


llm.invoke("What is LangChain?")

LangChain is a framework designed for developing applications...

Additional parameters: max_completion_tokens, temperature, etc.

¹ https://platform.openai.com/docs/quickstart

🤗 Prompting Hugging Face models

from langchain_huggingface import HuggingFacePipeline


llm = HuggingFacePipeline.from_model_id(
    model_id="meta-llama/Llama-3.2-3B-Instruct",
    task="text-generation",
    pipeline_kwargs={"max_new_tokens": 100}
)


llm.invoke("What is Hugging Face?")

Hugging Face is a popular open-source artificial intelligence (AI) library...

Let's practice!

Developing LLM Applications with LangChain