AgrowFit

AgriHome(Home page)
Services we provide(as we scroll down the home p
Live Soil Insights(Enter your district to get the required soil data provided by virtual sensors for suitable crop prediction)
Now fill in the soil data here
Our ml model will predict the most suitable crop based on the soil data entered
Predicted crop dislayed(Gram in this case)
Crop trends by region (click on any crop to know its detailed information )
Crop Transport finder(A map to find the crop transport service providers nearby along with their contact info)

Inspiration Every harvest season, thousands of farmers across Maharashtra gamble with uncertainty. Despite their hard work, many watch their crops fail—not due to lack of effort, but due to a lack of information. They plant what they hope will grow, not what’s best for their soil and climate. We met farmers who had lost an entire season's income because they didn’t know their soil lacked nitrogen or that the rain would come too late. That pain stuck with us. So we built AgroWfit—a solution born out of empathy and fueled by technology.

What it does AgroWfit is more than just a tool—it's a companion for the modern Indian farmer. It predicts the most profitable and suitable crop for any region in Maharashtra using real-time soil insights and weather data. Our platform empowers farmers to:

Get accurate crop recommendations (98% accuracy) using a custom-trained ML model

Access live soil data via virtual sensors mapped across districts

Use an interactive dashboard to visualize regional trends in pH, rainfall, temperature, and nutrient levels

Find nearby transportation services to efficiently move their crops post-harvest

Talk to our AI assistant for guidance on fertilizers, crop cycles, and best practices—like a digital krishi mitra (farm friend)

How we built it We began by diving into datasets(800+ samples)—soil health reports, weather patterns, district maps—and training a Random Forest model to make accurate predictions based on seven key soil features.

Then came the real challenge: turning data into something usable. We built a full-stack React app from scratch, created district-wise dashboards using Chart.js and Bing Maps, and embedded a live sensor map powered by Google My Maps.

To ensure security and ease of access, we added Google authentication, and to bridge the gap between tech and trust, we developed an AI chatbot that speaks the language of the farmer.

Challenges we ran into One of our biggest hurdles was making complex data feel simple. We weren't just building for coders—we were building for people who wake up at sunrise to till their fields. Making an intuitive, helpful, and mobile-friendly platform took iteration, testing, and listening to real users.

We also struggled to bring accuracy to our model across diverse districts—but with careful feature engineering and dataset augmentation, we pushed it to a confidence level we’re proud of.

Accomplishments that we're proud of Delivered 98% accurate crop prediction specifically tuned for Maharashtra’s soils

Created a real-time soil dashboard that’s both educational and easy to use

Connected transport logistics and personalized support in one seamless flow

Most importantly—we built something that can save lives and livelihoods

What we learned This journey taught us that technology can only be meaningful when it listens. Our farmers don’t need flashy apps—they need reliable help. Building AgroWfit reminded us that innovation isn’t just about new tools—it’s about new trust.

We also learned the importance of localization, simplicity, and design thinking when solving real-world problems.

What’s next for AgroWfit Our mission doesn’t end here.

We want to reach every farmer in India, especially those in underserved and remote areas

Add voice-based interaction in local languages

Include fertilizer and pest guidance, backed by AI

Partner with agriculture ministries to roll this out nationwide

And eventually, turn AgroWfit into a daily digital assistant for every next generation farmer—turning uncertainty into confidence, one crop at a time