KRISHIMITRA
Your digital friend in the field
Created on 12th January 2026
•
KRISHIMITRA
Your digital friend in the field
The problem KRISHIMITRA solves
KrishiMitra empowers farmers to transform their agricultural practices and income through digital connectivity. Farmers can instantly connect with thousands of peers across India to ask questions, share experiences, and learn best practices - whether it's the right time to plant wheat in Punjab or how to deal with pest issues in paddy fields, someone in the community has the answer. They can sell their produce directly to buyers through the marketplace, eliminating middlemen and getting 30-40% higher prices - a farmer in Maharashtra can now directly sell tomatoes to a buyer in Mumbai without losing money to multiple intermediaries. The real-time market price feature helps farmers check mandi prices instantly before selling, ensuring they never get cheated by local traders who quote lower rates. When crops show disease symptoms, farmers can simply take a photo and get AI-powered diagnosis within seconds, along with treatment recommendations - no need to wait days for an agricultural officer to visit. The expert consultation chat provides 24/7 access to agronomists for personalized advice on fertilizers, irrigation, crop rotation, and soil health. Weather alerts help farmers plan their activities better - they get notified about upcoming rainfall, storms, or heat waves, preventing crop damage and optimizing planting schedules. Farmers can discover government schemes they're eligible for and get step-by-step guidance on applying for subsidies, loans, and welfare programs they previously didn't know existed. For buyers and businesses, the platform provides direct access to fresh produce from verified farmers, ensuring quality and fair prices. Agricultural experts and students can use it as a knowledge-sharing platform to disseminate modern farming techniques and research findings. Overall, KrishiMitra makes farming safer by providing early disease detection, easier by digitizing traditionally manual processes like price discovery and market access, more profitable by connecting farmers directly to markets, and less isolating by building a supportive community where farmers can learn, grow, and prosper together.
Challenges I ran into
Challenges I Faced
During the hackathon, I ran into several technical roadblocks. Our real-time chat feature kept duplicating messages when multiple users were online simultaneously, and I couldn't figure out why Redux state was behaving strangely with Socket.io events. The ML disease detection API wouldn't connect to my frontend - I kept getting CORS errors no matter what I tried, which was frustrating because both services were running fine individually. Our MongoDB Atlas connection kept timing out and throwing authentication errors, making it impossible to test our database operations. When I tried to containerize everything with Docker, the containers wouldn't talk to each other and we wasted almost 2 hours troubleshooting networking issues. The scariest moment was 30 minutes before submission when our demo completely crashed during image upload testing - the entire app would freeze whenever someone tried to upload a crop photo, and I panicked thinking I wouldn't have a working demo to present.
How I Solved Them
I fixed the Socket.io duplication issue by carefully managing event listeners - I learned to properly remove listeners on component unmount and used unique message IDs to prevent duplicates, which took about 3 hours of debugging and reading documentation. For the CORS problem, I dove into FastAPI documentation and discovered I needed to add specific middleware configuration with allow_origins=["*"] and proper headers - something I hadn't done initially. The MongoDB timeout issue turned out to be simple - I just needed to whitelist our IP address in MongoDB Atlas network settings, which found after checking the error logs carefully. The Docker networking challenge was solved by creating a custom bridge network in docker-compose and ensuring all services were on the same network - a quick YouTube tutorial helped us understand this. For the last-minute crash, we quickly wrapped the image upload code in try-catch blocks and added fallback UI to show error messages gracefully instead of breaking the entire app. We also added client-side image compression to prevent large files from causing issues. These challenges taught us the importance of error handling, reading documentation thoroughly, and staying calm under pressure!
Tracks Applied (1)