Girl at school using SUrface Pen to write on tablet

Developing story: How I helped the next generation learn to code

Microsoft recently partnered with NASA and Acorn Aspirations to run a hackathon aimed at inspiring young people aged 11-18 to solve real-world problems using data science and artificial intelligence.

Over the three days, girls and boys designed, coded and took part in talks focused on AI and machine learning.

Kalai Elumalai is a Cloud Solution Architect at Microsoft and, along with his son Karthik, joined in with the hackathon at Microsoft’s London Reactor. Here, in his own words, is what happened.


I have organised and supported a number of data and AI-related hackathons but this was the first time I worked with young people as a Tech Mentor.

I took my 10-year-old son, despite having concerns that it might be too ambitious for him. However, he proved me wrong by actively participating with other youngsters during the app challenge.

A number of short sessions for tech experts helped the teams progress during the hackathon. The “design thinking” session, for example, provided the teams with a structured approach towards their chosen app challenge.

It was great for me to connect and collaborate with mentors from such a range of backgrounds, including entrepreneurs, business modellers and design thinking experts.

I was amazed by the youngsters’ knowledge of APIs and the use of machine learning to infuse intelligence in their apps. They saw AI as just another tool that could be used to augment their solution.

Many of the participants used Visual Studio Code as their integrated development environment for Python. and Cognitive Services such as Microsoft’s speech translator and the image classification API to augment their application.

They also researched their projects and created PowerPoints, and some wrote code for their apps, which included one on spotting wildfires. If a person sees a fire, he or she can take a photo and upload it to the app so others in the area can be aware of it, too. The app will also calculate the probability of the fire spreading, based on the weather data.