JeffKatzy/mvc-langchain

MVC Langchain pattern implementation.

• Separates models from workflows (ie. views), to allow for separate logic from parsing and prompting.
• Both sequenced and non-sequenced responses achieved via routing.
• When sequenced workflow used, looks at skip logic of each question to see if step is complete.

To run:

• git clone https://github.com/JeffKatzy/mvc-langchain
• cd mvc-langchain
• python3 -m venv venv
• pip3 install -r requirements.txt
• mv .env.example .env
• Add your openai api key
• python3 -i console.py

Todo:

• Potentially refactor more with langchain
  • Refactor with langgraph state? Does this slow it down?
• Still should look into Zep for output parsing
• Look at Cerebras for latency