I will build an ai agent workflow with langgraph and python

Need an AI workflow that does more than one prompt: reads documents, calls tools, checks its own output and hands back a result?

I build AI agents and multi-step workflows with LangGraph and Python.

What I have built:

- A document translation system with LangGraph agent nodes (planning, translation, review and refinement) for PDF and DOCX files, preserving layout

- A multi-stage LangGraph pipeline that extracts tasks from emails and documents, detects conflicts and removes duplicates

- MCP integrations with Gmail and Google Drive

What you get:

- A working LangGraph workflow in Python, with a diagram of the graph

- Tool calls to the APIs or services you need

- Setup instructions, and optionally a FastAPI endpoint

Please message me with what the agent should do, which tools or data it needs, and which LLM you use.