🚀 Personal Brand Oracle PRO

Hybrid AI/ML System for Predictive Content Strategy & Market Intelligence.

Personal Brand Oracle is a sophisticated SaaS-grade application designed to empower LinkedIn thought leaders. It doesn't just generate content; it audits and predicts performance using a unique combination of Classical Machine Learning and Generative AI.

---

🧠 The "Oracle" Architecture

What makes this tool unique is its Ensemble Evaluation System. Instead of relying solely on an AI's "opinion," every generated post passes through two distinct evaluation layers:

1. ML Auditor (Structural Layer): A RandomForestRegressor trained to analyze the "DNA" of a post. It evaluates physical features: scanability, paragraph density, hook questions, and engagement-driving characters.
2. LLM Judge (Semantic Layer): A Gemini 1.5 Flash agent acting as a Viral Marketing Expert. It evaluates copywriting quality, emotional resonance, and brand voice alignment.
3. The Hybrid Score: The final "Oracle Score" is a weighted average of both, providing a realistic prediction of how the LinkedIn algorithm and humans will react.

---

✨ Key Features

🔍 1. Market Intelligence Agent

• Real-time Gap Analysis: Uses DuckDuckGo Search to find what's trending and, more importantly, what's missing in your niche.
• Topic Discovery: Identifies 3 high-potential content "gaps" to position you as a pioneer.

⚖️ 2. A/B Simulator with Style Cloning

• Tone Synchronization: Optional TikTok link integration to extract and clone your specific writing style.
• Dual-Variant Generation: Automatically creates a Storytelling (Emotional) and a Technical (Authority) version of your idea.

📊 3. Predictive Analytics Dashboard

• Dynamic Scoring: Visual breakdown of structural vs. semantic quality.
• One-Click Copy: Built-in clipboard integration for seamless posting.

---

🛠️ Tech Stack

• Language: Python 3.10+
• Framework: Streamlit
• Orchestration: LangChain
• LLM: Google Gemini 1.5 Flash
• Machine Learning: Scikit-Learn (Random Forest)
• Search Engine: DuckDuckGo Search API

---

🚀 Installation & Setup

1. Clone the repository:

   git clone https://github.com/PaoloSebastian12/Personal_Brand_Oracle.git cd Personal_Brand_Oracle

2. Create a Virtual Environment:

   python -m venv brand

   Windows:

   brand\Scripts\activate

   Mac/Linux:

   source brand/bin/activate

3. Install Dependencies: pip install -r requirements.txt

4. Run the Dashboard: streamlit run app.py

🌐 Live Demo You can try the Personal Brand Oracle live here: 👉 Live Demo - Try the App here!