This is a complete, hands-on LangChain cheat sheet: install, configure, operate, automate and secure it - with copy-paste commands grouped by task. Bookmark it for day-to-day work, incidents and interviews alike.
LangChain is a framework for composing LLM applications: prompts, chains, tools, memory and retrieval-augmented generation (RAG), with LangSmith for tracing and evaluating agent behaviour.
Why you reach for it
- Reusable building blocks for LLM apps and agents
- First-class RAG and tool-calling patterns
- LangSmith tracing to debug and evaluate agents
Installation
pip install langchain langchain-openai langchain-community
export OPENAI_API_KEY=... # keep keys in a secret managerConfiguration reference
A tiny RAG chain over local documents.
from langchain_openai import ChatOpenAI, OpenAIEmbeddings
from langchain_community.vectorstores import FAISS
vs = FAISS.from_texts(docs, OpenAIEmbeddings())
retriever = vs.as_retriever(search_kwargs={'k': 4})
llm = ChatOpenAI(model='gpt-4o-mini', temperature=0)
def answer(q):
ctx = '\n'.join(d.page_content for d in retriever.invoke(q))
return llm.invoke(f'Use only this context:\n{ctx}\n\nQ: {q}').contentCore commands you use daily
print(answer('What is our deployment process?'))Setup & tracing
pip install langchain langchain-openai langchain-community langsmith
export OPENAI_API_KEY=...
export LANGCHAIN_TRACING_V2=true
export LANGCHAIN_API_KEY=... # LangSmithMinimal chain
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(model='gpt-4o-mini', temperature=0)
print(llm.invoke('Summarize GitOps in one sentence').content)Use it in CI/CD
Run evals against a golden set in CI with LangSmith.
- run: |
pip install langsmith
python evals/run.py # scores answers vs a labelled dataset, fails under thresholdBest-practice checklist
- Evaluate against a golden dataset in CI before shipping prompts
- Bound retrieval (top-k) and cache embeddings
- Scope agent tools with least privilege and human approval
Common pitfalls to avoid
- No evaluation, so prompt changes silently regress quality
- Unbounded context blowing token limits and cost
- Giving agents tools with broad, unguarded permissions
Security quick wins
- Keep API keys in a secret manager; never in code
- Guard tool use and validate model output before acting on it
Works well with
- Vector Databases - Similarity search over embeddings - the retrieval backbone of RAG and semantic search (Pinecone, Weaviate, Qdrant).
- MLflow - Open-source MLOps for experiment tracking, model registry and reproducible ML deployment.
- Python for DevOps - The glue language of DevOps - automation scripts, CLIs, API clients and operators.
A cheat sheet is muscle memory on paper - keep it close until the commands live in your fingers.
Want senior help putting LangChain to work - or a team bootcamp on it? CodeWizard runs Israel's largest DevOps community and embeds named senior engineers with a transparent hours log. Book a free consult or explore the Academy.