Editor's note: We're celebrating this year's impressive 20 Google Science Fair finalist projects over 20 days in our Spotlight on a Young Scientist series. Learn more about each of these inspiring young people and hear what inspires them in their own words.

Name: Girish Kumar
Home: Singapore
Age category: 16-18
Project: RevUp: improving learning through auto-generated study questions

Between receiving his first computer as a gift from his father and eight years of Robotic Club membership, Girish can’t remember a time when he wasn’t driven by science or building. As a student, he frequently relies on online texts to supplement classroom materials, but he doesn’t feel like they prepare him enough to learn relevant concepts. So Girish recruited help from local professors to code a tool that automatically generates questions spurred by manually inserted and relevant text. RevUp helps students study and provides an extremely efficient way for teachers to stop crafting their own review materials and start relying on automation.

What was the inspiration behind your project?
Last year, my entire class was obsessed with QuizUP, a mobile game where users could challenge friends to trivia questions. However, the more we played, the more repetitive questions started to become, taking the challenge and adrenaline rush away. This was because actual people were drafting questions, which limited the number of questions. That got me thinking: if there was way that questions could be automatically generated, players could get a much greater variety of questions. And of course, a more exciting game experience! I got hooked.

Soon after, my friend joked that if QuizUp was used in lessons, he would ace biology with great ease. Biology is an extremely knowledge-driven and content-driven subject in my school and quizzing often helps reinforce key concepts. That comment struck me, and I decided to “pivot” my idea to one that suited education.

When and why did you become interested in science?
Unlike most teenage programmers, I wasn’t fascinated with my first computer. I was disappointed.

As a kid, I used to watch countless TV shows that often depicted computers as intelligent, intelligent enough to engage in conversations. And obviously, the first computer my family bought was not able to do that. I was pretty disappointed. And that’s where and when my interest in Artificial Intelligence (AI) really took flight.

Things naturally progressed after that. I joined my elementary school’s robotics team, got into programming and took online courses on Machine Learning and AI in high school. I worked on exciting machine learning projects, dealing with topics such as gesture recognition and indoor localisation, as an intern at local laboratories. I felt a need to leverage A.I. to empower people. That’s why I’m continuing to work on RevUP: to put it in the hands of students and teachers worldwide.

What words of advice would you share with other young scientists?

  1. Do not waste time overthinking or over planning. Get your hands dirty and just start building/experimenting.
  2. For engineers especially: Iterate fast. Build. Fail. Improve. Repeat.
  3. Failure often paves way for insightful and surprising discoveries.