-
-
Logo
AgroLink - Overview
Inspiration
If you look closely at agricultural markets, one contradiction stands out.
Farmers wake up early, harvest their crops, and head to local markets — often without knowing what price they will receive. At the same time, another region may be facing shortages of the exact same crop, paying significantly higher prices.
The issue is not production.
The issue is coordination.
Most systems today operate in isolation:
- farmers depend on nearby mandis
- demand signals remain fragmented
- decisions are made without visibility
We built AgroLink to solve this disconnect — by creating a system where supply is guided by real demand signals rather than guesswork.
What it does
AgroLink is a demand-driven agricultural coordination platform that connects farmers with regional procurement needs through a structured intermediary system.
Instead of listing produce and waiting for buyers, farmers are shown where demand actually exists, based on real mandi price trends over time. When a region shows consistent price increase, it signals a demand-supply imbalance — and farmers can immediately respond by offering their produce.
Once a farmer submits an offer, the system moves it through a clear pipeline:
- request submission
- intermediary approval
- logistics coordination
- delivery tracking
At the same time, AgroLink includes an AI-powered advisory layer that helps farmers make better decisions by analyzing:
- price trends across markets
- optimal selling locations
- expected future price movement
- risk and timing
The result is a system where raw market data is transformed into actionable decisions.
How we built it
We approached AgroLink as a system problem rather than just an application.
The frontend was built using React and Vite, allowing us to create fast, responsive dashboards tailored for both farmers and intermediaries. The backend, powered by Node.js and Express, handles the core application logic and acts as a bridge between the interface and the AI layer.
At the core of the system is a simple but effective data model: we analyze mandi prices over a 7-day window to detect meaningful trends. These trends are converted into demand signals, which drive the entire supply flow.
On top of this, we integrated an AI layer using a large language model via Groq’s API. Instead of acting as a chatbot, the AI is structured to generate clear, decision-focused outputs — identifying the best market, predicting trends, and suggesting optimal selling windows.
The system is designed around a dual-role architecture:
- farmers interact with demand and supply
- intermediaries manage approval and logistics
This separation ensures both clarity and real-world alignment.
Challenges we ran into
One of the biggest challenges was designing a system that aligns with how agricultural supply chains actually function. It was tempting to build a direct farmer-to-buyer model, but in reality, intermediaries play a critical role in coordination, logistics, and validation.
Another challenge was building trust into the system. Farmers cannot rely on opaque predictions, so we ensured that all demand signals are grounded in transparent, verifiable data such as mandi price trends.
We also had to address the uneven distribution of supply and demand across regions. Moving supply efficiently requires not just visibility, but structured coordination — which is why we introduced intermediary-driven approval workflows.
Finally, we had to balance intelligence with simplicity. The system needed to provide meaningful insights without overwhelming the user, especially for those with limited technical exposure.
Accomplishments that we're proud of
We are proud of building a system that goes beyond a typical marketplace and reflects real-world agricultural processes.
AgroLink successfully:
- transforms price data into actionable demand signals
- creates a structured, end-to-end supply workflow
- integrates AI in a meaningful, decision-oriented way
- maintains simplicity while solving a complex coordination problem
Most importantly, it is not just a concept — it is a system that can realistically be extended into real-world deployment.
What we learned
Building AgroLink reinforced an important principle: solving real-world problems is less about complexity and more about clarity.
We learned that:
- simple, well-structured logic can outperform complex models
- real-world alignment matters more than theoretical design
- data becomes valuable only when converted into decisions
- user experience is critical, especially for non-technical users
It also showed us that technology’s role is not just to automate, but to enable better decision-making.
What's next for AgroLink
The current version establishes a strong foundation, but there is significant room for expansion.
Our next steps include integrating real-time mandi data sources to improve accuracy, enhancing logistics coordination through route optimization, and building a mobile-first interface tailored for farmers in the field.
We also plan to introduce reliability scoring systems to build trust within the network and expand the platform across multiple regions.
In the long term, AgroLink aims to evolve into a nationwide coordination layer — ensuring that agricultural supply flows efficiently to where it is needed most.
Mathematical Insight
Demand-supply imbalance can be interpreted as:
$$ \text{Demand Signal} \propto \frac{\text{Price Increase Over Time}}{\text{Supply Stability}} $$
Log in or sign up for Devpost to join the conversation.