OpenAI API & LangChain

Creating API keys, environment variables, quota, billing management

Token pricing, when cost matters, model selection strategy

Chains, agents, tools, memory — the abstraction layer explained

Loading PDFs, text files, web pages — LangChain document loaders

chunk_size, chunk_overlap, why recursive, keeping paragraphs together

LLM + retriever + memory, return_source_documents, verbose mode

End-to-end: load → split → embed → store → retrieve → answer

Why same question gets different answers, randomness, temperature effect