PsyDesign AI

Turn a single sentence into a complete, psychology-driven brand identity.

PsyDesign AI is an AI design partner that takes one idea and returns a full brand — a strategy report, a DALL·E 3 logo, font pairings, and a chat-ready brand persona — in seconds, on a high-performance FastAPI backend.

Overview · Features · Tech Stack · How It Works · Demo Notice · Getting Started · License

---

📖 Overview

PsyDesign AI automates the early branding process end to end. You describe a business in plain language; the platform uses GPT-4o to build a structured brand-psychology report, feeds the logo concept into DALL·E 3 for a unique mark, and spins up an interactive brand persona you can talk to — and listen to. Every result is saved so you can revisit, compare, or delete past brands.

Strategy, visuals, and voice — produced in a single pipeline.

---

✨ Features

🧠 Brand Psychology Report

A structured strategy document generated from one prompt:

• Personality — brand archetype (e.g. The Hero, The Sage), tone of voice, and core values
• Visual identity — a live Google Fonts pairing, ready for the web
• Names & slogans — multiple unique brand-name and tagline options
• Narrative — a brand story plus a ready-to-post social bio
• Audience — a detailed target persona with demographics, interests, and pain points
• Marketing hook — an opening campaign concept to kick-start strategy

🎨 Logo Generation (DALL·E 3)

• Builds a precise image prompt from the report's logo_concept_description
• Produces HD 1024×1024 logos — minimalist, iconic, and versatile
• One-click download through a secure image-proxy endpoint

💬 Interactive Brand Persona

• Chat with your brand — a GPT-4o persona that embodies the new brand's tone
• On-topic by design — politely declines anything unrelated to the brand
• Text-to-speech — hear responses via OpenAI TTS-1-HD

🌐 Platform

• 6 languages — English, Spanish, French, Arabic, Chinese, and Hindi, for both UI and AI output
• Persistent history — every brand (report + logo) stored via SQLAlchemy
• Responsive UI — mobile-first, clean, and animated

---

🛠 Tech Stack

Layer	Technology	Role
Backend	Python 3.10+	Core logic & async request handling
	FastAPI	High-performance, type-safe API with auto OpenAPI docs
	SQLAlchemy	ORM for database access
	Pydantic	Request / response validation
	Uvicorn	ASGI server
AI	GPT-4o	Brand report + chat persona
	DALL·E 3	Context-aware logo generation
	TTS-1-HD	Natural-sounding voice for the persona
Frontend	JavaScript (ES6+)	Client logic, dynamic rendering, async calls
	HTML5 & CSS3	Structure and responsive, animated design
Database	SQLite	Local history storage (swappable for PostgreSQL)

---

🏗 How It Works

flowchart LR
    A[User prompt] --> B[FastAPI endpoint<br/>Pydantic validation]
    B --> C[GPT-4o<br/>Brand report JSON]
    C --> D[DALL·E 3<br/>Logo from concept]
    C --> E[(SQLite<br/>save history)]
    D --> E
    E --> F[Frontend<br/>renders report, logo,<br/>fonts & chat]

Loading

1. Input — the user submits a brand description.
2. Validate — FastAPI receives the request; Pydantic checks the data.
3. Strategy — a structured JSON prompt goes to GPT-4o, returning the brand report.
4. Logo — the report's logo_concept_description becomes a DALL·E 3 prompt → logo URL.
5. Persist — report JSON + logo URL are saved as a history item via SQLAlchemy.
6. Render — the frontend parses the response and builds the results page, including font previews and the live chat window.

---

⚠️ Demo Notice

This repository is a demo build of PsyDesign AI.

• OpenAI API calls are replaced with mock data.
• The production API key and internal optimization scripts are removed for security.
• Full source and proprietary models are available under a separate purchase agreement.

For the full version or IP transfer, contact artinghorbanianainex@gmail.com.

---

🏁 Getting Started

Prerequisites

• Python 3.10+
• Git
• An OpenAI API key (only needed for live generation)

1 — Clone

git clone https://github.com/YourUsername/PsyDesign-AI.git
cd PsyDesign-AI

2 — Create a virtual environment (recommended)

# macOS / Linux
python3 -m venv venv
source venv/bin/activate

# Windows
python -m venv venv
.\venv\Scripts\activate

3 — Install dependencies

Create a requirements.txt in the project root:

fastapi
uvicorn[standard]
sqlalchemy
openai
httpx
jinja2

Then install:

pip install -r requirements.txt

4 — Add your API key (optional)

The demo runs on mock data with no key. To enable live generation, open app.py and set your key (around line 20):

# app.py
API_KEY = "YOUR_OPENAI_API_KEY"

Security: in production, load the key from an environment variable or a secrets manager rather than hardcoding it.

5 — Run

uvicorn app:app --reload

Open http://127.0.0.1:8000 in your browser. 🚀

---

📜 License

Copyright © 2025 Artin Ghorbanian. All Rights Reserved.

This project and its source code are proprietary and confidential. Unauthorized copying, distribution, or use of this software, via any medium, is strictly prohibited.