ExpSolver

Multi agent architecture to evaluate the BODMAS expressions

View project on GitHub

Multi-Agent BODMAS Calculation Engine

A production-grade, highly resilient calculation engine using LangChain, LangGraph, and FastMCP with full Langfuse observability.

Overview

This system evaluates raw string mathematical expressions (e.g., 10 รท (3 + 2)) using a state-reduction loop powered by 5 independent agents (Planner, Adder, Subtractor, Multiplier, Divider). It uses a local Ollama endpoint by default, with graceful fallbacks to OpenAI or Anthropic API providers.

System Prerequisites

  • Python 3.10+
  • Docker & Docker Compose (for PostgreSQL & Langfuse)
  • Optional: Local Ollama instance running

Quick Execution Recipe

  1. Clone the repository.
  2. Setup virtual environment: 
    python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
  3. Configure Environment: 
    cp .env.example .env
# Edit .env with your specific API keys if not using Ollama
  4. Boot Infrastructure (Langfuse & PostgreSQL): 
    docker-compose up -d
  5. Run the MCP Server: 
    ./scripts/run_mcp.sh

Telemetry Trace Navigation

All executions are traced via Langfuse.

  • Open http://localhost:3000
  • Login and view the generated traces to see exactly how the multi-agent graph evaluated your mathematical expressions step-by-step.