# `LangchainPrompt.Profiles`
[🔗](https://github.com/exfoundry/langchain_prompt/blob/v0.1.0/lib/langchain_prompt/profiles.ex#L1)

Indirection layer for named AI execution profiles.

Delegates `get/1` to the module configured under
`config :langchain_prompt, :profiles_impl, MyApp.AIProfiles`.

This allows applications to define their own named profiles without
coupling prompt modules to a concrete profile module.

## Setup

1. Create a module that implements `get/1`:

       defmodule MyApp.AIProfiles do
         alias LangchainPrompt.Profile
         alias LangchainPrompt.Adapters.Langchain

         def get(:fast) do
           %Profile{
             adapter: Langchain,
             opts: %{chat_module: LangChain.ChatModels.ChatGoogleAI, model: "gemini-2.0-flash-lite"}
           }
         end
       end

2. Configure it:

       # config/config.exs
       config :langchain_prompt, :profiles_impl, MyApp.AIProfiles

       # config/test.exs
       config :langchain_prompt, :profiles_impl, LangchainPrompt.Profiles.TestImpl

3. Use it in prompt modules:

       def set_profile(_assigns), do: LangchainPrompt.Profiles.get(:fast)

# `get`

Returns a `Profile` for the given name, delegating to the configured implementation.

---

*Consult [api-reference.md](api-reference.md) for complete listing*