SiliconSage

SiliconSage is an AI-powered PC building assistant that optimizes computer builds for value and performance using machine learning.

Overview

A hybrid application combining a Next.js frontend with a Python (FastAPI) ML backend. It goes beyond compatibility checking to provide real-time performance analytics and architectural advice.

Key Features

• PC Builder & Visualizer: Interactive drag-and-drop interface with schematic visualization.
• ML Performance Lab: Real-time FPS prediction, bottleneck detection, and build integrity analysis (PSU efficiency & Motherboard tier checks).
• AI Advisor: Expert persona ("SiliconSage") powered by Llama 3 for context-aware part recommendations and troubleshooting.
• Ecosystem Comparison: Analytical tool comparing custom builds against consoles (PS5/Xbox) and laptops to ensure optimal value.

Technology Stack

Frontend

• Framework: Next.js 16 (App Router)
• Language: TypeScript
• Styling: Tailwind CSS
• Animations: Framer Motion
• Icons: Lucide React

Backend (ML Engine)

• Framework: FastAPI (Python)
• ML Libraries: Scikit-Learn, Pandas, NumPy
• Server: Uvicorn

Infrastructure

• Database: Supabase (PostgreSQL)
• AI Inference: Groq API (Llama 3)

Setup Instructions

Prerequisites

• Node.js 18+
• Python 3.10+
• A Supabase project
• A Groq API Key

1. Repository Setup

Clone the repository:

git clone https://github.com/sammy200-ui/SiliconSage.git
cd SiliconSage

2. Frontend Setup

Navigate to the frontend directory and install dependencies:

cd frontend
npm install

Create a .env.local file in the frontend directory:

NEXT_PUBLIC_SUPABASE_URL=your_supabase_url
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_supabase_key
GROQ_API_KEY=your_groq_api_key

Run the frontend development server:

npm run dev

The application will be available at http://localhost:3000.

3. ML Engine Setup

Open a new terminal and navigate to the ml_engine directory:

cd ml_engine

Create a virtual environment and activate it:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

Install Python dependencies:

pip install fastapi uvicorn scikit-learn pandas numpy joblib

Run the ML Engine server:

uvicorn main:app --reload --host 0.0.0.0 --port 8000

The API will be available at http://localhost:8000.

Architecture Notes

The Frontend and ML Engine are designed to run concurrently. The frontend makes direct API calls to localhost:8000/predict/fps to fetch real-time analytics. Ensure both servers are running for the full application experience.