Agents

RoutePlanner AI

Describe your stops in plain language and get an optimized route with turn-by-turn directions.

The problem

Sales reps, couriers and field teams plan multi-stop routes by hand and lose hours driving in the wrong order.

How it works

A six-node LangGraph flow parses the request with an LLM, geocodes each place, builds a distance matrix with Google Maps, solves the visiting order as a traveling-salesman problem with OR-Tools, and returns step-by-step directions.

Pipeline

01Parse
02Geocode
03Distance matrix
04Optimize (TSP)
05Directions

Highlights

Natural-language input — no forms
Route order solved as a TSP with OR-Tools
Single-responsibility nodes sharing typed state
Self-documented FastAPI endpoints

Stack

LangGraphOpenAIGoogle Maps APIsOR-ToolsFastAPI

Want something like this for your company?

I build production versions of these systems on your data and your infrastructure.

Related service: Agents & workflow automation

Book a call