AI / LangGraph LangChain Interview questions
How do you install and set up LangChain?
LangChain is distributed as several pip packages. The minimal install for OpenAI-backed applications is:
pip install langchain langchain-openai # For community integrations (vector stores, loaders, etc.): pip install langchain-community # For serving with REST API: pip install langserve fastapi uvicorn
API credentials are passed through environment variables so they never appear in source code:
export OPENAI_API_KEY="sk-..." # OpenAI export ANTHROPIC_API_KEY="..." # Anthropic / Claude export LANGCHAIN_TRACING_V2="true" # Enable LangSmith tracing export LANGCHAIN_API_KEY="ls__..." # LangSmith API key export LANGCHAIN_PROJECT="my-project" # LangSmith project name
A minimal "hello world" with LangChain:
from langchain_openai import ChatOpenAI from langchain_core.messages import HumanMessage llm = ChatOpenAI(model="gpt-4o-mini") response = llm.invoke([HumanMessage(content="What is 2 + 2?")]) print(response.content) # "4"
The package split is intentional: langchain-core contains stable base abstractions, langchain contains orchestration logic, langchain-openai and similar provider packages contain model integrations, and langchain-community contains third-party integrations that move faster.
Which environment variable enables LangSmith tracing?
Which package contains stable base abstractions like the Runnable interface?
More Related questions...
