Cropsense

Inspiration

The inspiration I got from is Agriculture in India, It is a backbone of India, about Approximately 54.8% of India's land is considered agricultural land, 55% of Indian population primarily dependent on agriculture in that 61% of India's farmers depend on rain-fed agriculture. which is highly depended on climate, Indias climate also very much changes if you go from south to north you can find tropicals, desserts and even snow mountains in this 3000kms. So this is my small contribution towards them to give farmers in India by analyzing climate in that particular longitude and latitude suggesting which crop that they should go for.

What it does

It will retrieve weather data and soil data for that particular pincode of the user in India and suggest him the best crop that he should go with, if he very much depended on climate

How we built it

We used AWS EC2 for hosting our code, we used Bedrock as an LLM by AWS to power our Agent by Strandsagent which also supported by AWS, First we need to convert pincode to longitude and latitude, and then we used open-meteo API which is free of cost for past climate data, and openepi API which is open source community project which provides free api on soil data, they also contained much more..

so we got climate data and soil data.

I have done web scrapping and created a dataset of ideal weather conditions required for top 100 crops in india and I have assigned embeddings to them

Next I have to find best crop with the embedding the is closely matching to previously fetched climate data

Challenges we ran into

In India, agriculture data is not easily available, it was very difficult to find ideal conditions to crops.

Accomplishments that we're proud of

The usecase that we have chosen, which has a lot of scope in India to help farmers, agriculture lenders in finance, vegetable sellers in supply chain, FMCG companies. the scale, the Impact, that considerably matters to the 130CR population of country

What we learned

we have learned how to use AWS services, learned a lot about crops, seasons, farmers and many more about our country

What's next for Cropsense

It can be used for giving insights to farmers, their are also major banks that gives agriculture loans to farmers, farmers may not have eligibilities like civil scores, we can use this tool for agriculture lending,

we can add market price analysis to it, so it can also consider highly variable prices so that farmers can plan about warehouse storage properly.