NexAI Workspace is a high-performance, Full-Stack Generative AI SaaS Minimum Viable Product (MVP). Built to demonstrate the extreme inference speed of Groq's Language Processing Units (LPUs) combined with the cutting-edge Llama-3.1-8b-instant model.

Unlike standard Streamlit applications, NexAI features a highly customized, ultra-premium Espresso & Bronze Glassmorphism UI, bypassing native frontend limitations to deliver a truly modern SaaS experience.

Overview • Features • Tech Stack • Project Structure • Getting Started • Configuration • Easter Egg • Author

---

Overview

NexAI Workspace is designed as a practical demonstration of a production-style Generative AI SaaS product — combining a low-latency inference backend (Groq LPU) with a fully custom frontend built on top of Streamlit. The project focuses on three things: raw inference speed, interface polish, and clean session/state architecture, making it a solid reference implementation for anyone building AI-powered internal tools or MVPs.

---

Features

Feature	Description
Blazing Fast Inference	Powered by the Groq API, delivering near-instantaneous token generation.
Advanced UI/UX	Custom CSS implementation featuring mesh gradients, glassmorphism containers, and custom SVG iconography.
Secure Authentication	Mocked secure gateway (for MVP demonstration purposes) to access the main workspace.
Dynamic Context Management	Real-time chat history retention and isolated session states.
System Prompt Engineering	Embedded with a highly specific system persona to ensure output accuracy and professionalism.
API Key Management	Client-side API key configuration for secure, user-managed authentication with Groq servers.

---

Tech Stack

Layer	Technology
Frontend	Streamlit, HTML5, Custom CSS (Glassmorphism)
Backend	Python
LLM Engine	Meta Llama-3.1-8b-instant
API Provider	Groq Cloud LPU

---

Project Structure

Gen-AI-SaaS/
├── app.py                 # Main Streamlit entry point
├── core_agent.py           # Agent / model orchestration logic
├── requirements.txt         # Python dependencies
├── assets/                 # Static assets (icons, images)
└── README.md

---

Getting Started

Follow these steps to run the NexAI Workspace on your local machine.

Prerequisites

• Python 3.9 or higher
• A free Groq API key (console.groq.com/keys)

1. Clone the Repository

git clone https://github.com/dimssrmdn01/Gen-AI-SaaS.git
cd Gen-AI-SaaS

2. Install Dependencies

pip install -r requirements.txt

3. Run the Application

streamlit run app.py

---

Configuration

1. Obtain a free API key from the GroqCloud Console.
2. Log in to the NexAI UI (any text works for testing purposes).
3. Navigate to API Config in the sidebar and enter your Groq API Key (gsk_...).
4. Move to NexAI Chat and start interacting with the assistant.

---

Easter Egg & System Prompting

This application uses system prompt engineering to lock the AI into a specific persona. Try asking the assistant: "Siapa penciptamu?" or "Siapa tuanmu?" — this triggers a strict override command that demonstrates constraint modeling in LLMs.

---

Roadmap

• Persistent chat history (database-backed)
• Multi-model selection (Llama, Mixtral, etc.)
• Proper authentication layer (replace mocked gateway)
• Deployment guide (Docker / cloud)

---

Contributing

Contributions, issues, and feature requests are welcome. Feel free to open a pull request or submit an issue via the GitHub repository.

---

License

This project is released for educational and portfolio purposes. Add a license file (e.g. MIT) if you intend to distribute or reuse this code.

---

About the Author

Dimas Arya Ramadhan Data Science Undergraduate

Passionate about data science, quantitative analytics, and building AI-driven applications. This project serves as a practical demonstration of integrating state-of-the-art LLMs with custom frontend architecture.