Wazihub Soil Moisture Prediction Challenge
$8,000 USD
Predict soil humidity using sensor data from low-cost DIY Internet of Things in Senegal
29 July–20 October 2019 23:59
699 data scientists enrolled, 96 on the leaderboard
Auto ML
published 13 Oct 2019, 15:58

Can one use Auto ML, since the AutoML algorithm will handle the choice of the model given the field data, but the user will not change hyperparamters of the autoML across all fields. Even though it's not really the same model for all the fields, it's a good fit for the problem of different fields and the final user doesn't have to tune any hyper parameter.

I believe AutoML is discouraged. You don't end up with code to submit and it's expensive past the free tier (meaning those with more money can pay for better models). Where AutoML is fun is to get a benchmark to aim for - once you know what is possible it's much easier to gear up and try to do better :)