Distributed research platform

VinQ (https://vin-q.com) aims to find scientifically validated solutions to initiate a global transition to more sustainable practices in agriculture through participative research.

Aim

Initiate the shift from traditional towards more sustainable, nature centered regenerative practices in agriculture using data-centered approach.

Actionable intelligence

This system uses AI to analyze and learn from individual actions, which are then cross-referenced with real-world consequences, facts and outcomes. This demonstrates the use of AI being trained on real-world data to improve human decision-making as an advocate of Nature, rather than on human-labeled training data.

How it works

Data-centric knowledge sharing

1. Collaboration based on sharing of reliable data. The farmers collect data individually in their fields and indicating their priorities and personal objectives.

2. Two classes are connected with AI:

• Data from Nature: True, verifiable data reflecting the actual state (ground truth) such as microbiology, climate, soil quality, diseases, and production, which can be confirmed through fact-checking. This data represent points in a parameter space. Any changes in this data are reflected.

• Actions The other type of data pertains to the management style inducing the changes in the state, which includes actions that affect points in a parameter space. Any treatments or fieldwork carried out to mitigate problems or improve conditions reflect the personalized style.

By utilizing AI to connect these two types of data, a governance system is established, where the user autonomously explore parameter space while AI is trained on real-world consequences thus providing the farmer a feedback on the consequences for the Nature of his actions.

Dashboard

Activities

Previous funding:

This project has not received any centralized grants or funding. It received donations through Gitcoin rounds: Climate rounds of GC18, GC19, Metagov, Hyperdesci: $189.86 crowdfunded from 134 people $609.96 received from matching pools.

Hypercerts are minted.

The money was used to build sensors and NAYAD remote sensing online platform for prediction of downy mildew and collection of basic fields in Penedes wine region. This setup is scalable.