The LangChain
ecosystem
D E V E L O P I N G L L M A P P L I C AT I O N S W I T H L A N G C H A I N
Jonathan Bennion
AI Engineer & LangChain Contributor
Meet your instructor...
Jonathan Bennion, AI Engineer
ML & AI at Facebook, Google, Amazon,
Disney, EA
Created Logical Fallacy chain in LangChain
Contributor to DeepEval
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Build LLM Apps with LangChain
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
LangChain integrations
1 [Link]
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Building LLM apps the LangChain way...
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Building LLM apps the LangChain way...
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Building LLM apps the LangChain way...
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Building LLM apps the LangChain way...
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Building LLM apps the LangChain way...
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Prompting OpenAI models
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
model="gpt-4o-mini",
api_key='...'
)
[Link]("What is LangChain?")
LangChain is a framework designed for developing applications...
Additional parameters: max_completion_tokens , temperature , etc.
1 [Link]
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
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}
)
[Link]("What is Hugging Face?")
Hugging Face is a popular open-source artificial intelligence (AI) library...
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Let's practice!
D E V E L O P I N G L L M A P P L I C AT I O N S W I T H L A N G C H A I N
Prompt templates
D E V E L O P I N G L L M A P P L I C AT I O N S W I T H L A N G C H A I N
Jonathan Bennion
AI Engineer & LangChain Contributor
Prompt templates
Recipes for defining prompts for LLMs
Can contain: instructions, examples, and additional context
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Prompt templates
from langchain_core.prompts import PromptTemplate
template = "Expain this concept simply and concisely: {concept}"
prompt_template = PromptTemplate.from_template(
template=template
)
prompt = prompt_template.invoke({"concept": "Prompting LLMs"})
print(prompt)
text='Expain this concept simply and concisely: Prompting LLMs'
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
llm = HuggingFacePipeline.from_model_id(
model_id="meta-llama/Llama-3.3-70B-Instruct",
task="text-generation"
)
llm_chain = prompt_template | llm
concept = "Prompting LLMs"
print(llm_chain.invoke({"concept": concept}))
Prompting LLMs (Large Language Models) refers to the process of giving a model a
specific input or question to generate a response.
LangChain Expression Language (LCEL): | (pipe) operator
Chain: connect calls to different components
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Chat models
Chat roles: system , human , ai
from langchain_core.prompts import ChatPromptTemplate
template = ChatPromptTemplate.from_messages(
[
("system", "You are a calculator that responds with math."),
("human", "Answer this math question: What is two plus two?"),
("ai", "2+2=4"),
("human", "Answer this math question: {math}")
]
)
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Integrating ChatPromptTemplate
llm = ChatOpenAI(model="gpt-4o-mini", api_key='<OPENAI_API_TOKEN>')
llm_chain = template | llm
math='What is five times five?'
response = llm_chain.invoke({"math": math})
print([Link])
5x5=25
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Let's practice!
D E V E L O P I N G L L M A P P L I C AT I O N S W I T H L A N G C H A I N
Few-shot prompting
D E V E L O P I N G L L M A P P L I C AT I O N S W I T H L A N G C H A I N
Jonathan Bennion
AI Engineer & LangChain Contributor
Limitations of standard prompt templates
PromptTemplate + ChatPromptTemplate
examples = [
Handling small numbers of examples {
"question": "..."
Don't scale for many examples
"answer": "..."
FewShotPromptTemplate
},
...
]
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Building an example set
examples = [
{
"question": "Does Henry Campbell have any pets?",
"answer": "Henry Campbell has a dog called Pluto."
},
...
]
# Convert pandas DataFrame to list of dicts
examples = df.to_dict(orient="records")
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Formatting the examples
from langchain_core.prompts import FewShotPromptTemplate, PromptTemplate
example_prompt = PromptTemplate.from_template("Question: {question}\n{answer}")
prompt = example_prompt.invoke({"question": "What is the capital of Italy?"
"answer": "Rome"})
print([Link])
Question: What is the capital of Italy?
Rome
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
FewShotPromptTemplate
prompt_template = FewShotPromptTemplate(
examples=examples,
examples : the list of dicts
example_prompt=example_prompt,
suffix="Question: {input}", example_prompt : formatted template
input_variables=["input"] suffix : suffix to add to the input
)
input_variables
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Invoking the few-shot prompt template
prompt = prompt_template.invoke({"input": "What is the name of Henry Campbell's dog?"})
print([Link])
Question: Does Henry Campbell have any pets?
Henry Campbell has a dog called Pluto.
...
Question: What is the name of Henry Campbell's dog?
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Integration with a chain
llm = ChatOpenAI(model="gpt-4o-mini", api_key="...")
llm_chain = prompt_template | llm
response = llm_chain.invoke({"input": "What is the name of Henry Campbell's dog?"})
print([Link])
The name of Henry Campbell's dog is Pluto.
DEVELOPING LLM APPLICATIONS WITH LANGCHAIN
Let's practice!
D E V E L O P I N G L L M A P P L I C AT I O N S W I T H L A N G C H A I N
