Speaker
Description
Hack4dev, was launched in May 2022 by Prof. Carolina Odman, with the participation of several partners, including the Inter-University Institute for Data Intensive Astronomy (IDIA), the Office of Astronomy for Development (OAD), the BRICS Intelligent Telescope and Data Network (BITDN), Development in Africa with Radio Astronomy (DARA), and the African Astronomical Society (AfAS). Since then, it has grown into a global initiative that connects research and development through hackathons. It serves as a shared backbone for hackathons, offering models, guidance, and support while welcoming contributions of tools, resources, and infrastructure from around the world.
As artificial intelligence (AI), machine learning (ML), and data science become increasingly important across research, industry, and development sectors, access to practical training opportunities, mentorship, and applied learning experiences remains uneven, particularly in emerging innovation ecosystems. The Hack4dev Data Science Hackathon Programme (DSHP) was developed to address this challenge through a scalable, decentralised approach to technical skills development.
The programme employs a two-stage cascade model designed to expand access to high-quality data science training while fostering local ownership and innovation. Selected organisers first participate in a Trainers Hackathon, where they receive training in technical challenge facilitation, community engagement, and hackathon management. These organisers subsequently implement Regional Hackathons within their own institutions and communities, creating locally driven learning environments while contributing to a shared global challenge framework.
Findings from the programme’s first cycle demonstrate the effectiveness of this approach. More than 50 organisers supported over 150 participants across multiple regions, resulting in 48 submitted machine learning and data science solutions developed within a three-day period. Twenty submissions outperformed the benchmark model provided by the organisers, while the highest-performing solution achieved execution speeds 165 times faster than the baseline. These outcomes demonstrate that short-term, challenge-driven hackathons can simultaneously strengthen technical competencies and encourage peer-to-peer learning.
Building on these outcomes, the 2025–2026 programme cycle will expand to 22 hackathons hosted across institutions in Botswana, Ethiopia, Ghana, Iraq, Kenya, Madagascar, Mozambique, Namibia, Nigeria, South Africa, Tanzania, Zambia, and Zimbabwe. This expansion provides an opportunity to evaluate the scalability and reproducibility of AI and data science capacity-building models across diverse educational and institutional contexts. A forthcoming third programme cycle will further extend Hack4dev's reach by recruiting new organisers and institutions, creating additional opportunities for hands-on machine learning training, mentorship, and community-led innovation. The programme aims to strengthen regional data science ecosystems while promoting inclusive participation and collaboration across the world.
| Apply for student award at which level: | None |
|---|---|
| Consent on use of personal information: Abstract Submission | Yes, I ACCEPT |