Aurum AI🧠

Aurum AI is a cutting-edge personal finance and investment assistant designed to empower retail investors. It combines a suite of powerful tools to provide personalized financial insights and recommendations.

🔍 Overview

At Mumbai Hacks 2024, we tackled the finance domain within the investment track, creating an assistant that seamlessly integrates AI-driven insights for enhanced investment strategies. Aurum AI's primary features include:

1. Phi 3.5-Based Real-Time RAG Chat Assistant
   A conversational AI chat assistant answers personal finance questions in real-time, using Phi 3.5 to retrieve and generate accurate responses.

2. Mutual Fund Recommendation Engine
   Leveraging a machine learning model trained on a custom-scraped dataset of 14,000 data points, this engine provides personalized mutual fund recommendations based on user requirements.

3. IPO Prediction Neural Network
   Our custom black box neural network, trained on a meticulously updated dataset of 300 data points, achieves 100% accuracy in identifying viable IPOs for investment.

---

🛠️ Tech Stack

We utilized the following technologies to build Aurum AI:

• Backend: Django, FastAPI, SQLite
• Frontend: HTMX, Tailwind CSS
• Machine Learning: TensorFlow, Keras, Scikit-learn, Statsmodels, Spacy
• Data Processing: Pandas, NumPy

---

🚀 Features

• Real-Time Chat: Seamlessly answer finance-related queries.
• Personalized Investment Insights: Tailored mutual fund suggestions based on 14k+ data points.
• Accurate IPO Predictions: A neural network providing reliable IPO recommendations with unmatched precision.

---

⚙️ Getting Started

To set up and run Aurum AI, follow these steps:

1. Environment Setup

1. Clone this repository:

   git clone github.com/Aurum-MumbaiHacks2024/aurum
   cd aurum-ai

2. Set up a virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

2. Install Backend Dependencies

pip install -r requirements.txt

3. Frontend Setup

npm install
npm run build:css
npm run dev

4. Start the Backend API Services

• Run FastAPI for API endpoints:

  uvicorn app:app --reload

• Alternatively, use FastAPI's run command:

  fastapi run app.py

• Run Django for additional backend services:

  python manage.py runserver

---

🎯 Future Plans

• Expand Financial Services: Add advisory features tailored for venture capitalists (VCs), offering investment insights and portfolio guidance.
• Commercialization: Introduce a subscription-based payment model, allowing users to access advanced analytics and premium support.

---

🤝 Contributors

• Varad Joshi
• Meet Jamsutkar
• Rohan Waghode
• Vaishnavi Saarang

For feedback or collaboration, reach out to us at varad,[email protected].

---