StubbleX

• 

  Connecting farmers and industries through intelligent biomass trade.

Inspiration

Every year, we witness severe air pollution caused by stubble burning after harvest seasons. While media and policymakers often blame farmers, the reality is much more complex. Farmers burn crop residue not by choice, but because they lack affordable, accessible alternatives. At the same time, industries such as bio-energy plants and paper mills actively search for sustainable raw materials. This gap — waste on one side and demand on the other — inspired us. We asked ourselves: What if agricultural waste could become a verified, tradable asset instead of an environmental liability? That question led to the creation of StubbleX.

What it does

StubbleX is an AI-powered biomass marketplace that connects farmers directly with industries and transforms crop residue into economic and environmental value. The platform allows farmers to: List crop waste for sale Use AI-based image classification to identify residue Receive dynamic price predictions Calculate environmental impact Estimate carbon credit value Industries can: Browse verified biomass listings Track procurement and carbon offset metrics Generate sustainability reports We also introduced: A Voice-First Multilingual Assistant that allows farmers to list waste using local dialects. A Carbon Credit Calculator that estimates CO₂ emissions avoided and translates them into potential carbon credit revenue. The CO₂ savings are calculated using: CO₂ Saved = Tons of Stubble × Emission Factor per Ton This transforms waste reduction into measurable climate impact.

How we built it

StubbleX was built as a full-stack web application with two primary portals: 1.Farmer Portal 2.Industry Portal Core components include: -Computer Vision model for waste classification -AI-based price estimation logic -Environmental impact modeling engine -Carbon credit calculation module -Voice-to-text multilingual assistant The frontend was designed for simplicity and rural accessibility, while the backend handles listing management, AI inference, pricing logic, and marketplace transactions. We focused on creating a working MVP that demonstrates: -Real-time listing -AI-assisted decisions -Procurement tracking -Impact visualization

Challenges we ran into

One of our biggest challenges was designing a solution that was both technologically advanced and rural-friendly. We faced challenges in: -Making AI outputs understandable for farmers -Designing a UI simple enough for non-tech users -Structuring carbon impact calculations realistically -Simulating demand-based pricing logic Another challenge was ensuring that the solution felt scalable and not just like a prototype. Balancing innovation with practicality was key.

Accomplishments that we're proud of

We are proud that StubbleX is not just a concept — it is a functional working prototype. We successfully built: -A dual-portal marketplace -AI-based waste identification -Dynamic pricing estimation -Voice-enabled multilingual listing -Carbon credit calculation system -Environmental impact tracking dashboard Most importantly, we created a solution that combines: -Farmer empowerment -Climate action -AI innovation -Economic sustainability

What we learned

Through building StubbleX, we learned: -Technology must be accessible to create real impact. -AI is powerful only when it solves human problems. -Sustainability solutions must also be economically viable. -Climate impact can be quantified and monetized through proper modeling. We also learned how to think beyond just building features — and instead build systems.

What's next for StubbleX

Our next steps include: -Integrating real-time demand forecasting models -Partnering with bio-energy companies for pilot deployment -Adding logistics route optimization -Implementing verified carbon credit certification workflows -Expanding language support for rural regions Our long-term vision is to scale StubbleX beyond one region and create a global AI-powered agricultural climate platform.

Built With

• css
• fastapi
• framer
• github
• motion
• next
• python
• react
• render
• scikit-learn
• tensorflow
• typescript
• vercel
• xgboost