1. Document ingestion
   - Parse documents
   - Split into chunks
   - Generate embeddings

2. Index creation
   - Store in vector DB
   - Add metadata

3. Query pipeline
   - Embed query
   - Retrieve relevant chunks
   - Rerank results

4. Response generation
   - Assemble context
   - Generate response
   - Add citations

1. Data preparation
   - Collect examples
   - Format for model
   - Split train/eval

2. Training
   - Upload dataset
   - Configure hyperparameters
   - Start job

3. Evaluation
   - Run on eval set
   - Compare to base model
   - Check for regressions

4. Deployment
   - Deploy model
   - Monitor performance
   - Iterate

from openai import OpenAI

client = OpenAI()

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ],
    temperature=0.7,
    max_tokens=1000
)

print(response.choices[0].message.content)

import anthropic

client = anthropic.Anthropic()

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)

print(message.content[0].text)

from openai import OpenAI
import chromadb

# Create embeddings
client = OpenAI()
embedding = client.embeddings.create(
    model="text-embedding-3-small",
    input="Search query"
)

# Store and search
chroma = chromadb.Client()
collection = chroma.create_collection("docs")

collection.add(
    embeddings=[embedding.data[0].embedding],
    documents=["Document text"],
    ids=["doc1"]
)

results = collection.query(
    query_embeddings=[query_embedding],
    n_results=5
)

from langchain_openai import ChatOpenAI, OpenAIEmbeddings
from langchain_chroma import Chroma
from langchain.chains import RetrievalQA

# Setup
llm = ChatOpenAI(model="gpt-4o")
embeddings = OpenAIEmbeddings()
vectorstore = Chroma(embedding_function=embeddings)

# RAG chain
qa = RetrievalQA.from_chain_type(
    llm=llm,
    retriever=vectorstore.as_retriever(),
    return_source_documents=True
)

result = qa.invoke({"query": "What is RAG?"})

# OpenAI caching example
response = client.chat.completions.create(
    model="gpt-4o",
    messages=[
        {
            "role": "system",
            "content": long_system_prompt,  # Cached after first call
        },
        {"role": "user", "content": user_query}
    ]
)