GramAI is an AI-powered decision-support platform designed to help rural entrepreneurs evaluate and plan a business using hyper-local market intelligence, financial analysis, government scheme matching, risk assessment, and personalized AI guidance in one place.
The platform is designed around a simple principle:
Data → Intelligence → Advice
Instead of relying on a chatbot alone, GramAI combines deterministic business logic with AI so that important calculations and decision signals are generated by the backend, while the AI Advisor explains the results in a simple conversational way.
Rural entrepreneurs often have limited access to timely and contextual business guidance. Important decisions such as:
- What business should I start?
- Is there enough local demand?
- How much competition exists nearby?
- How much money will I need?
- How much should I invest myself?
- How much loan may be required?
- What will the EMI look like?
- Which government schemes may apply?
- What are the major business risks?
are often answered using fragmented information or informal guidance.
GramAI brings these areas together into a single platform.
The user provides a location, business category, and financial requirements. GramAI processes this information through multiple intelligence engines and produces a structured business analysis.
- Hyper-local market intelligence
- Local competition analysis
- Market Map visualization
- Financial and loan planning
- EMI and repayment calculation
- Business viability scoring
- Business risk assessment
- Government scheme matching
- Applicant-profile based scheme personalization
- AI-powered Business Advisor
- English, Hindi and Hinglish conversational interaction
- Voice input
- RAG-lite knowledge retrieval
- Structured business reports
- PDF report generation
- Real-time analysis progress using SSE
- Persistent analysis and report history
GramAI evaluates the selected business within its local context.
The market analysis can consider:
- Estimated customer base
- Local demand
- Market opportunity
- Competition density
- Pricing pressure
- Selected location
- Business category
The goal is to move from generic business advice to location-aware business intelligence.
GramAI analyzes nearby competition within the selected market area.
The prototype uses controlled/demo competitor data to demonstrate:
- Number of competitors
- Competition density
- Nearby business locations
- Market competition signals
This information is also visualized through the Market Map.
The Market Map provides a visual representation of the selected business location and nearby competitors.
It helps the entrepreneur understand:
- Selected business location
- Nearby competitors
- Approximate market area
- Competition distribution
- Local market signals
The current prototype uses a custom map-style visualization with controlled/demo location data.
GramAI provides structured financial planning based on the entrepreneur's requirements.
The financial engine handles calculations such as:
- Project cost
- Own contribution / margin
- Loan requirement
- Interest rate
- Loan tenure
- Moratorium
- EMI
- Repayment planning
- Financial feasibility
The core financial calculations are performed by the backend financial engine rather than being generated by the LLM.
GramAI combines multiple decision signals to produce a business viability assessment.
The decision logic considers factors such as:
- Market demand
- Competition
- Financial feasibility
- Business risk
The resulting Viability Score is intended to give the entrepreneur a simple high-level signal before making further business decisions.
Risk intelligence provides a structured view of potential operational, market and financial risks.
GramAI matches business requirements and applicant information with configured government scheme criteria.
The system can use applicant attributes such as:
- Woman entrepreneur
- SC/ST category
- Previous Tarun loan repayment
- Business/sector type
- Project cost
- Loan requirement
The prototype provides preliminary scheme matches and displays relevant eligibility information and benefits.
Government scheme eligibility should always be verified against the latest official guidelines before making an application.
The AI Advisor provides conversational guidance based on the business analysis.
Users can ask questions about:
- Business feasibility
- Pricing
- Competition
- Financing
- EMI and loans
- Government schemes
- Market demand
- Business risks
- Working capital
- Viability results
The Advisor uses the available analysis context so that responses can be related to the entrepreneur's specific business scenario.
GramAI is designed to make business guidance easier to access for users who may prefer regional or conversational communication.
The current prototype supports:
- English
- Hindi
- Hinglish-style interaction
- Voice input through the browser Speech Recognition API
The production roadmap includes broader regional-language support and more complete voice-first interaction.
The AI Advisor can use a lightweight retrieval layer to provide relevant contextual information.
The current prototype uses a RAG-lite / controlled knowledge approach, rather than a full production-scale vector database and document-ingestion pipeline.
The production architecture can be extended with:
- Official government documents
- Scheme guidelines
- Market datasets
- Business knowledge
- Embeddings
- Vector database
- Source and date tracking
GramAI converts the analysis into a structured business report containing information such as:
- Viability
- Market analysis
- Competition
- Financial structure
- Pricing recommendations
- Risk assessment
- SWOT analysis
- Government scheme matches
- Key assumptions
Reports can be generated as PDF documents for further planning and verification.
Rural Entrepreneur
|
Voice / Text / Input
|
v
React + TypeScript Frontend
|
REST API / SSE
|
v
FastAPI Backend
|
+----------------+----------------+
| | |
v v v
Market / GIS Financial Scheme / RAG
Engine Engine Engine
| | |
+----------------+----------------+
|
v
Decision Engine
+-------------+-------------+
| | |
Demand Financial Competition
| | |
+-------------+-------------+
|
v
Viability & Risk
|
v
AI Advisor
LLM + Context
|
v
Report Generator
|
v
PDF Report
Deterministic systems calculate. AI explains and personalizes.
Financial calculations, scheme matching logic, competition estimation and viability signals are handled by backend services and decision logic.
The LLM is used primarily for natural-language interaction, contextual explanation and personalized guidance.
- React.js
- TypeScript
- Vite
- CSS
- Axios
- Python
- FastAPI
- Uvicorn
- Pydantic
- SQLite
- SQLAlchemy
- JWT-based authentication
- LLM-based AI Advisor
- RAG-lite / contextual retrieval
- Rule-based and deterministic business logic
- Market Intelligence Engine
- Competition Analysis Engine
- Financial / Loan Engine
- Government Scheme Matching Service
- Viability Engine
- Risk Assessment Logic
- Browser Speech Recognition API
- Hindi / English language recognition
- ReportLab
- PDF generation
- Server-Sent Events (SSE)
- Custom Market Map
- Controlled/demo location and competitor data
GramAI/
|
+-- backend/
| +-- app/
| | +-- api/
| | +-- agents/
| | +-- finance/
| | +-- services/
| | +-- db.py
| | +-- main.py
| |
| +-- requirements.txt
| +-- ...
|
+-- frontend/
| +-- src/
| | +-- components/
| | +-- pages/
| | +-- dashboardApi.ts
| | +-- App.tsx
| +-- package.json
| +-- ...
|
+-- docs/
| +-- screenshots/
| +-- dashboard.png
| +-- market-map.png
| +-- schemes.png
| +-- advisor.png
| +-- report.png
|
+-- .gitignore
+-- README.md
Make sure the following are installed:
- Python 3.11
- Node.js LTS
- npm
- Git
Recommended development environment:
- Windows
- VS Code
- Conda environment
git clone https://github.com/ridamgupta79-dev/GramAI.git
cd GramAIActivate the project environment:
conda activate gramaiInstall backend dependencies:
pip install -r backend/requirements.txtMove into the frontend directory:
cd frontendInstall dependencies:
npm installCreate a local environment file if required by the configured AI services:
.env
Do not commit secrets or API keys to GitHub.
A shareable environment template can be maintained as:
.env.example
From the project root:
uvicorn app.main:app --reload --app-dir backendThe backend will be available at:
http://localhost:8000
Swagger API documentation:
http://localhost:8000/docs
Open another terminal:
cd frontend
npm run devThe frontend will normally be available at:
http://localhost:5173
1. User selects location
|
2. User selects business category
|
3. User enters financing requirements
|
4. GramAI starts analysis
|
5. Market intelligence
|
6. Competition analysis
|
7. Financial feasibility
|
8. Scheme eligibility matching
|
9. Viability & risk assessment
|
10. AI advisory insights
|
11. Dashboard results
|
12. PDF business report
The analysis pipeline provides stage-wise progress using Server-Sent Events.
The backend is organized around API modules for major platform capabilities.
Key areas include:
/api/auth
/api/v1/dashboard
/api/analysis
/api/reports
/api/schemes
/api/locations
The exact available endpoints may evolve as the prototype is developed further.
The current prototype uses a controlled/demo data approach to demonstrate the complete GramAI workflow.
Data and inputs include:
- User-provided business information
- User-selected locations
- Seeded location information
- Controlled competitor/market scenarios
- Configured government scheme information
- Application-level business rules
- Financial inputs
The prototype does not currently represent a complete live national data integration layer.
Potential production data integrations include:
- Government scheme sources
- Agmarknet and market data
- OpenStreetMap / geospatial data
- NABARD resources
- RBI financial information
- Other verified public datasets
These integrations are part of the future production roadmap.
- React + Vite frontend
- FastAPI backend
- User authentication flow
- Location selection
- Business analysis workflow
- Hyper-local market analysis
- Competition analysis
- Market Map
- Financial calculations
- Loan planning
- EMI calculation
- Viability Score
- Business risk assessment
- Government scheme matching
- Applicant-profile based scheme matching
- AI Advisor
- English / Hindi / Hinglish interaction
- Voice input
- RAG-lite contextual retrieval
- SSE analysis progress
- PDF report generation
- Reports page
- Persistent analysis/report history
- Dashboard restoration after refresh
The current version is a functional prototype and has several limitations.
- Market and competitor information is controlled/demo data.
- Live Agmarknet integration is not currently implemented.
- Live government API integration is not currently implemented.
- National-scale datasets are not yet integrated.
- The current knowledge retrieval layer is RAG-lite.
- A production-scale embedding/vector database pipeline is not yet implemented.
- Fully offline/local AI is not currently implemented.
- Multilingual support can be expanded further.
The current prototype focuses on financial structuring and feasibility, including:
- Project cost
- Own contribution
- Loan amount
- Interest
- Tenure
- EMI
- Repayment planning
A complete day-to-day financial ledger and transaction-record system is part of the future roadmap.
- SQLite is used for the prototype.
- Production deployment can migrate to PostgreSQL.
- The current analysis worker is designed for prototype use.
- Production-scale distributed processing can use dedicated workers and infrastructure.
The following are future integrations:
- WhatsApp / Telegram
- Banking integrations
- Offline synchronization
- Live government and market data
- Production cloud infrastructure
- Working business analysis
- Market intelligence
- Competition analysis
- Financial planning
- Scheme matching
- AI Advisor
- Voice input
- Market Map
- PDF reporting
- Live market data integration
- Verified government scheme knowledge base
- Full RAG pipeline
- Vector database
- Source/date tracking
- Better geospatial intelligence
- Improved viability models
- Full voice-first workflow
- Regional language expansion
- Offline-first architecture
- Lightweight/local AI models
- Low-bandwidth optimization
- Voice-based financial records
- WhatsApp / Telegram support
- Banking and financing integrations
- Application assistance
- Business progress tracking
- Financial identity building
- Large-scale cloud deployment
GramAI is designed as a decision-support system, not a replacement for professional financial, legal or government-authority verification.
AI-generated recommendations may contain errors.
For important decisions:
Record → Prompt → Verify → Decide
Financial calculations should be checked against the user's actual financial terms.
Government scheme eligibility should be verified against the latest official scheme guidelines before application.
Market insights are dependent on the quality, coverage and freshness of the underlying data.
GramAI aims to become a rural-first business intelligence platform that helps entrepreneurs move from:
Business Idea
|
Local Market Understanding
|
Financial Planning
|
Risk & Viability Assessment
|
Government Scheme Discovery
|
Personalized AI Guidance
|
Actionable Business Plan
The long-term vision is to make reliable business intelligence and financial guidance more accessible to rural entrepreneurs, even in environments with limited connectivity, limited digital literacy and limited access to professional advisory services.
Ridam Gupta
AI/ML-focused developer building GramAI as a rural business intelligence and decision-support platform.
GramAI — AI-Powered Hyper-Local Business Advisory & Financial Intelligence
Built as an AI/ML and full-stack project focused on rural entrepreneurship, business intelligence, financial planning and accessible decision support.
All project screenshots are stored in:
docs/screenshots/
Current screenshot files:
dashboard.png
market-map.png
schemes.png
advisor.png
report.png
This project is currently developed as a hackathon/prototype project.