company-os

CompanyOS ๐Ÿข โ€” Autonomous Company Engine

CompanyOS is an autonomous company operating system that runs ANY company with a team of 7 executive AI agents.

Given a single company.json configuration file, CompanyOS stands up a fully functioning company, generates an operating plan (mission โ†’ department goals โ†’ backlog tasks), assigns tasks to executive agent roles, tracks execution on an interactive Kanban board, streams live agent activity logs, and updates real-time financial KPIs.


โšก Key Features


๐Ÿš€ Quickstart Guide

1. Requirements

2. Run Locally

# Navigate to company_os directory
cd /Users/venkataswaraswamy/company_os

# Run the app
python3 app.py

Open your browser and navigate to: ๐Ÿ‘‰ http://localhost:5000


๐ŸŽฏ Pointing CompanyOS at ANY Company

Edit company.json in the root directory:

{
  "name": "Acme SaaS Inc.",
  "industry": "B2B Cloud Analytics",
  "description": "Real-time telemetry and predictive analytics platform for mid-market engineering teams.",
  "products": ["Analytics Cloud", "Telemetry SDK", "Enterprise Alerting"],
  "target_market": "VP of Engineering & DevOps Managers",
  "team_size": 25,
  "location": "San Francisco, CA"
}

Then click ๐Ÿ”„ Rebuild Plan in the dashboard UI or execute:

curl -X POST http://localhost:5000/api/rebuild

CompanyOS will dynamically regenerate the entire company plan, department goals, and task backlog tailored specifically to your company profile!


๐Ÿ”‘ Running in LIVE Mode (Optional)

To enable live LLM intelligence for all 7 executive agents:

export OPENROUTER_API_KEY="your-openrouter-key"
export OPENROUTER_MODEL="deepseek/deepseek-v4-flash"  # Optional, default is deepseek/deepseek-v4-flash
python3 app.py

๐Ÿงช Running Unit Tests

Run the offline pytest suite to verify agent logic, plan generation, board transitions, and KPI calculations:

python3 -m pytest test_companyos.py -v

๐Ÿ“ Directory Architecture

company_os/
โ”œโ”€โ”€ app.py               # Flask application & HTTP REST API endpoints
โ”œโ”€โ”€ company.json         # Company configuration file
โ”œโ”€โ”€ company_os/
โ”‚   โ”œโ”€โ”€ __init__.py
โ”‚   โ”œโ”€โ”€ config.py        # Config loader & defaults
โ”‚   โ”œโ”€โ”€ agents.py        # 7 Executive AI Agent classes (SIM + LIVE think)
โ”‚   โ”œโ”€โ”€ engine.py        # Plan generation, Assign loop, Board advance & KPIs
โ”‚   โ””โ”€โ”€ store.py         # JSON storage & append-only event logger
โ”œโ”€โ”€ data/                # Data storage directory (plan, board, kpis, events)
โ”œโ”€โ”€ static/
โ”‚   โ”œโ”€โ”€ style.css        # Dark theme dashboard styles
โ”‚   โ””โ”€โ”€ app.js           # Frontend interactivity & live polling
โ”œโ”€โ”€ templates/
โ”‚   โ””โ”€โ”€ index.html       # Dashboard HTML template
โ”œโ”€โ”€ test_companyos.py    # Offline unit test suite
โ”œโ”€โ”€ requirements.txt     # Python dependencies
โ””โ”€โ”€ README.md

Author & Enterprise AI Architecture Hub

Built by Nagarjuna Reddy Kamireddy (Nagarjuna Kamireddy / Nagarjuna K) โ€” Lead AI Architect, GenAI & Agentic AI Lead, and Corporate AI Trainer (20+ years across Enterprise Data, SAP HANA & Production AI; AWS Certified Solutions Architect โ€“ Professional & NVIDIA-Certified Professional: Agentic AI).