Skip to content

KavyaSambarelly/FinAdvisor-project

Repository files navigation

FinAdvisor-project

FinAdvisor project details and files as well as the complete project pdf file

💬 Digital Financial Literacy AI Agent

An AI-powered assistant designed to help users understand and navigate digital finance safely. Built using IBM Watsonx.ai and Retrieval-Augmented Generation (RAG), this intelligent agent provides multilingual support and trustworthy guidance on UPI, interest rates, stock market, personal finance, investing , loan safety, fraud prevention, online scams and more.


🧩 Problem Statement

Many individuals, especially from rural or digitally underserved areas, lack financial literacy. They struggle with understanding UPI, online scams, interest rates, or personal budgeting. This can lead to fraud, poor money management, or digital exclusion.


💡 Proposed Solution

An AI agent trained using IBM Watsonx.ai that delivers reliable, document-based answers about digital finance. With multilingual support and document retrieval (RAG), it ensures users receive grounded, easy-to-understand financial guidance.


🧠 Technologies Used

  • IBM Watsonx.ai Studio
  • IBM Granite Foundation Model (LLM)
  • Vector Index for Retrieval-Augmented Generation
  • PDF documents from RBI/NPCI & other related sourses
  • NLP (Natural Language Processing)
  • IBM Cloud Object Storage

☁️ IBM Cloud Services Used

  • Watsonx.ai Studio
  • IBM Granite Model
  • Watsonx Vector Index
  • IBM Cloud Lite Account
  • IBM Cloud IAM
  • IBM Cloud Object Storage

👥 End Users

  • General public seeking financial clarity
  • Rural and semi-urban citizens
  • Students and young professionals
  • First-time UPI and digital banking users
  • NGOs and government outreach programs
  • Customer service centers
  • Self-Help Groups / Women’s Collectives
  • Educators / Institutions

🌟 WOW Factors

  • Uses RAG to fetch answers from real RBI/NPCI or uploaded PDFs
  • Built entirely on IBM Cloud using Watsonx tools
  • Handles unrelated questions with polite redirection
  • Future-ready with multilingual support
  • Educates users on scams, interest rates, budgeting, stock market, UPI, and more
  • Provides safe and personalized financial guidance

🧪 Key Features

  • Document-based Q&A via Vector Index
  • Powered by IBM Granite LLM for natural language understanding
  • Built-in NLP support
  • Responds gracefully to off-topic/irrelated questions
  • Covers real financial concerns (scams, loans, savings, UPI, etc.)

🚀 How It Works

  1. User inputs a query (e.g., “How to avoid UPI fraud?”)
  2. IBM Granite LLM processes the language
  3. Vector Index retrieves trusted content from uploaded financial PDFs
  4. Agent responds with a grounded, natural-language answer

📌 How to Run or Deploy

  1. Log in to IBM Cloud Lite: https://cloud.ibm.com
  2. Launch Watsonx.ai Studio
  3. Create a new AI Agent
  4. Upload financial PDFs to a Vector Index
  5. Choose Tools for web search (Google,Wikipedia,DuckDuckGo etc..)
  6. Configure agent instructions and topics (restricting AI from answering off-topic questions politely)
  7. Test in the preview panel
  8. Deploy via web snippet, Streamlit, or custom web UI

🛣️ Future Scope

  • WhatsApp or mobile app integration
  • Speech-to-text input for voice-driven queries
  • Automatic monthly report generation
  • Region-specific financial policy updates
  • Multilingual expansion with Watson Language Translator

🔗 Useful Links


🔗 Connect with me on LinkedIn


Created with 💙 during the IBM SkillsBuild for Academia Internship 2025 by Kavya Sambarelly

About

FinAdvisor project details and files as well as the complete project pdf file

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published