Arkitect 

Arkitect is an AI-powered workflow planning platform that transforms a simple goal into a structured execution roadmap.

Users can enter goals such as:

"Launch a SaaS startup in 90 days"

and Arkitect generates:

• Deliverables
• Tool Recommendations
• Step-by-Step Workflows
• AI Prompts
• Knowledge Areas
• Alternative Approaches
• Estimated Timelines

The platform combines AI-generated planning with a deterministic fallback pipeline to ensure reliable workflow generation even when external AI services are unavailable.

---

📸 Screenshots

Landing Page

LOGIN

Generated Workflow

---

✨ Features

AI Workflow Generation

Generate structured execution plans from natural language goals.

Deliverable Detection

Automatically identifies key outputs required to achieve the objective.

Tool Recommendations

Suggests relevant tools, platforms, and technologies for execution.

AI Prompt Generation

Creates prompts that can be directly used with LLMs such as ChatGPT and Gemini.

Knowledge Area Identification

Highlights concepts and skills that should be learned during execution.

Alternative Strategies

Provides alternative workflow approaches and execution paths.

Time Estimation

Generates realistic effort and timeline estimates.

Deterministic Fallback Engine

If Gemini is unavailable or rate-limited, Arkitect automatically falls back to a deterministic workflow generation pipeline.

---

🏗 Architecture

User
 │
 ▼
Frontend (React + TanStack Start)
 │
 ▼
Backend API (FastAPI)
 │
 ├── Gemini AI Engine
 │
 └── Deterministic Workflow Engine
          │
          ▼
      PostgreSQL

---

🛠 Tech Stack

Frontend

• React 19
• TypeScript
• TanStack Start
• TanStack Router
• Vite
• Tailwind CSS
• Shadcn UI
• Supabase Auth

Backend

• FastAPI
• Python
• SQLAlchemy
• Pydantic
• Async PostgreSQL

Database

• PostgreSQL (Neon)

Authentication

• Supabase Authentication

AI

• Google Gemini API

Deployment

• Vercel (Frontend)
• Render (Backend)
• Neon PostgreSQL
• Supabase

---

📂 Project Structure

Arkitect/
│
├── frontend/
│   ├── src/
│   ├── routes/
│   ├── components/
│   └── lib/
│
├── backend/
│   ├── app/
│   │   ├── api/
│   │   ├── services/
│   │   ├── database/
│   │   └── auth/
│   │
│   └── tests/
│
├── docs/
│   ├── PRD.md
│   ├── DESIGN.md
│   ├── SYSTEM_CONTRACT.md
│   └── OWNERSHIP.md
│
└── README.md

---

⚙️ Local Setup

Clone Repository

git clone https://github.com/AlfinMuhammedS/Arkitect.git
cd Arkitect

---

Backend Setup

cd backend

python -m venv venv

# Windows
venv\Scripts\activate

# Mac/Linux
source venv/bin/activate

pip install -r requirements.txt

Create:

backend/.env

DATABASE_URL=
SUPABASE_URL=
SUPABASE_ANON_KEY=
SUPABASE_JWT_SECRET=
GEMINI_API_KEY=

Run backend:

python -m uvicorn app.main:app --reload

Backend available at:

http://localhost:8000

Swagger Docs:

http://localhost:8000/docs

---

Frontend Setup

cd frontend

npm install

Create:

frontend/.env

VITE_SUPABASE_URL=
VITE_SUPABASE_ANON_KEY=
VITE_API_URL=http://localhost:8000

Run frontend:

npm run dev

Frontend available at:

http://localhost:3000

or

http://localhost:5173

depending on configuration.

---

🧪 Testing

Backend tests:

cd backend

pytest

---

📈 Future Improvements

• Real-time workflow streaming
• Team collaboration
• Public workflow sharing
• Workflow export (PDF/Markdown)
• Advanced analytics
• Multi-model AI support
• Workflow templates marketplace

---

👨‍💻 Authors

👨‍💻 Authors

• Alfin Muhammed S https://github.com/AlfinMuhammedS
• Aadil Sandeep https://github.com/Aadilsandeep

---

📜 License

This project was developed for educational and portfolio purposes.