🏆 Best Business Impact Winner - NSA Boston National Hackathon 2025
Recognized for creating the most impactful solution addressing real-world agricultural challenges in Nepal.
Krisi Sahayak is a web application developed by Greenwise.ai to assist farmers in Nepal. Utilizing Google Cloud APIs, it offers voice and text-based agricultural guidance in Nepali and local dialects, providing real-time weather updates and localized crop recommendations.
To empower Nepalese farmers with a voice-driven AI assistant that delivers real-time market and weather insights, helping them increase profitability while collecting crucial agricultural data for future planning and analysis.
To create the first central repository of agricultural data in Nepal, empowering:
- Farmers with personalized advice
- NGOs with ground realities
- Policymakers with actionable insights
- 🗣️ Voice & Text Interaction: Communicate in Nepali or local dialects using Google Cloud's Speech-to-Text and Text-to-Speech APIs.
- 🤖 AI-Powered Recommendations: Receive personalized crop and yield advice powered by Google Gemini LLM.
- 🌦️ Weather Updates: Access location-based weather forecasts for informed farming decisions.
- 📊 Data Logging: Store user interactions and responses in Google Cloud for future analysis.
- 🖥️ User-Friendly Interface: Navigate a simple, visual dashboard designed for farmers.
- Frontend: Django Templates (HTML, CSS)
- Backend: Django + REST API
- AI Integration: OpenAI ChatGPT LLM
- Cloud Services: Google Cloud Speech-to-Text, Text-to-Speech, Cloud Storage
- Version Control: Git, GitHub
git clone https://github.com/prasiddhigyawali/NSAHackathon2025.git
cd NSAHackathon2025
-
Navigate to the project directory:
cd nsa_hackathon -
Create virtual environment:
python -m venv . -
Activate virtual environment:
# Windows: source Scripts\activate # Linux/Mac: source bin/activate
-
Install dependencies from requirements.txt:
pip install -r requirements.txt
-
Navigate to Django project:
cd greenwise -
Run Django server:
python manage.py runserver
(Optional for now) Set up Google Cloud credentials:
set GOOGLE_APPLICATION_CREDENTIALS="path/to/service-account.json"
python manage.py runserver
3️⃣ Frontend Access
Access the application via local server http://127.0.0.1:8000/.
-
Ethnic Language Model: Expand to create the first AI model specifically trained for Nepali ethnic languages and dialects, ensuring truly localized voice and text interaction.
-
Mobile Application: Offline-capable version for broader accessibility.
-
NGO/INGO and Government Partnerships: Collaborate to scale impact and insights.
🌱 Adishri Pradhan – Team Lead
🌱 Drishya Shrestha - Software Engineer
🌱 Prabesh Sunar - Business Analyst
🌱 Prasiddhi Gyawali - Software Engineer