FruitPunch AI for Good Challenge

by FruitPunch AI B.V.

Crowd-sourcing your SDG AI solutions

AI for Good Challenges

NOTE: For any questions, feel free to reach out to:

In an AI for Good Challenge, we crowd-source ~50 engineers from our AI for Good Community with a diverse set of cultural and professional backgrounds. This group works together in teams to develop a functioning AI prototype for your organization. Our engineers go from data to a solution in 10 weeks.

We manage the AI for Good Challenge from start to finish. During this time, we will host open events like master classes and final presentations, and generate media coverage and social media content.

After preparing the Challenge and gathering all the necessary info and data, we spent 6 weeks promoting the Challenge, setting up the infrastructure, and assembling the group of engineers that will solve your problem. We screen all participants and divide them into subteams, which get their own deliverables and a project coach. We host a plenary kick-off meeting and weekly meetings per subteam. All teams are in tight communication with our Data Scientists.

What we need

As the organization that provides the technical problem, you are the Challenge Owner. To be able to solve your problem, we’ll need a couple of things.

  1. A description of the problem that you are trying to solve.
    • Explain, first of all, why it is relevant to solve this problem. You’re probably down in the weeds; everything is quite clear to you, but not to us. Take a moment to reflect and explain the greater purpose of what you’re doing. Towards which Sustainable Development Goals would solving this problem contribute?
    • An explanation of how you will objectively measure success. What would solving this problem look like? What metrics or KPI’s will improve?
    • Next, get specific. Tell us what the problem is you want to solve with AI. Could you describe a (sub-)problem you are currently working on, or approaches you have tried?
  2. A description of the data that you have to solve the problem. You might need to reach out to your domain or tech experts here.
    • What kind of data do you have on the problem? Describe the type of data and the quantity.
    • Are the prediction targets labeled as part of the dataset?
    • Where does the data reside and how do you process it now? On notebooks on individual computers? On Azure or Amazon Web Services?
    • Is this data property of an organization or publicly available? If it’s not public: could this data be shared? With participants from any country?
  3. Dedication. One person will be the designated Challenge Owner. We need them to actively participate throughout the challenge. This means attending all events that we host. Since you’re the owner of this problem, you should be open to questions from the engineers working hard to solve it. If possible, your experts give the Domain Masterclass and Tech Masterclass.
  4. Communication. Partnering up means spreading the word of this AI for Good Challenge together. If you’re not able to reach a large group of aspiring AI engineers, find at least three partners who do. This could be the campus of your startup or lab, academic organizations, research institutes, interest groups, foundations or meetups. Maybe you’re in touch with a media partner who might publish an article on our collaboration. Someone in your team might have a large following on Linkedin—don’t be afraid to get creative and hussle, we sure aren’t.

Let’s apply AI for Good!

If you’ve got a problem, if you think AI is the answer, and if solving it would do Good,
don’t hesitate to reach out!

At a glance