The objective of this challenge was to create machine learning models that use open-
source CO2 emissions data (from Sentinel-5P satellite observations) to predict carbon
emissions. The overall aim was to bring teams from different areas of EY together to build
individual and team data science and analytics capability. This year, 76 participants
enrolled and 327 submissions were made.
Kavi Pather, Artificial Intelligence and Advanced Analytics Leader for EY Africa, said, “We’re
pleased to say that our partnership with Zindi has allowed many EY teams to upskill in the
critical areas of analytics and data science. Congratulations to all our winners and everyone
who took part. We are looking forward to a bigger and better competition in the future!”
The top 3 teams were celebrated at a hybrid event held in Johannesburg and Cape Town.
They are:
1st Place: Jacques Bisschoff, Christiaan Wessels & Justin Logie
2nd Place: Gerrit Dreyer & Daniel Ryklief
3rd Place: Hans Christian Von Stockhausen, Troy McNamara & Kamil Singh
This competition was hosted in partnership with Zindi (zindi.africa), Africa’s largest network
of data scientists. The competition was hosted on their platform, which provided EY
participants with access to the real-world data and automated evaluation system needed to
develop machine learning models to predict carbon emissions. These solutions will enable
EY, governments, and other actors to estimate carbon emission levels across Africa, even in
places where on-the-ground monitoring is not possible.
Celina Lee, CEO and co-founder at Zindi, said, “It was a pleasure working with Kavi Pather
and the incredible team at EY to launch their very own private AI challenges on Zindi. You
are a team doing things differently, building community and breaking down traditional silos!
Congratulations to the winning teams!”
