Inspiration
I was inspired to create Kisan Sahayak after learning about the challenges faced by farmers in developing countries like India. Farming is often overlooked, with many farmers having limited access to crucial information, struggling to understand complex documents, and having limited opportunities for training. I wanted to build a solution to bridge these gaps and empower farmers with the knowledge and resources they need to thrive.
What it does
Kisan Sahayak is a digital platform that provides farmers with:
- Easy access to information: Government schemes, helpful organizations, and contact details for experts.
- Educational resources: Provides modern farming techniques and best practices via Krishi Pathshala (Agriculture School).
- Document assistance: Clear summaries and explanations of complex documents.
How we built it
I built Kisan Sahayak using the following technologies:
- Frontend: Streamlit
- AI and Vector Search: Snowflake Cortex Search
- Large Language Model (LLM): Mistral AI (
mistral-large2) - Hosting: Streamlit Community Cloud
I used Streamlit to create a user-friendly interface, Snowflake Cortex Search to power the AI and search functionality, and Mistral-large2 to generate responses and summaries. The app is hosted on Streamlit Community Cloud for easy accessibility.
Challenges we ran into
Some of the challenges I encountered during development included:
- Data collection and cleaning: Gathering and preparing accurate and relevant data for the different sections of the app.
- Multilingual support: Ensuring accessibility and user-friendliness across multiple languages.
- Integrating different technologies: Seamlessly connecting Streamlit, Cortex Search, and Mistral-large2.
Accomplishments that we're proud of
I am proud of creating a functional and user-friendly app that addresses real-world challenges faced by farmers. I was able to successfully integrate different technologies to create a comprehensive solution. I am also proud of the app's multilingual support, which makes it accessible to a wider audience.
What we learned
I learned a lot about building RAG applications using tools and products offered by Snowflake, working with the Mistral model, and deploying apps on Streamlit Community Cloud. I also gained a deeper understanding of the challenges faced by farmers and the potential of technology to empower them.
What's next for Kisan Sahayak
I plan to continue improving Kisan Sahayak by:
- Adding more features and educational resources.
- Expanding language support to include more languages.
- Collaborating with agricultural organizations and experts to enhance support for farmers.
Built With
- cortex-search
- langchain
- mistral-ai
- python
- snowflake
- streamlit
Log in or sign up for Devpost to join the conversation.