How African Developers Are Contributing to Global Open Source Projects
For years, the story of software in Africa was told as a story of consumption. Developers downloaded the frameworks, libraries, and operating systems built elsewhere and shipped products on top of them. The record now looks different. A growing number of African developers write, review and maintain open source code, and a few governments have started to build public infrastructure on it.
The shift has been visible for some time. GitHub’s 2019 Octoverse report found that contributions across Africa were up 40%, more than on any other continent, with growth highest in Nigeria, Kenya, Tunisia and Morocco. The question in 2026 is how much of that early growth has turned into sustained participation.
A developer base that keeps expanding
A report by OpenUK, drawing on GitHub data, counts about 9.3 million developer accounts in Africa in the first quarter of 2026. Nigeria accounts for 1.77 million of them, up from 1.32 million a year earlier. Kenya rose from 463,595 to 666,020, and Rwanda from 55,098 to 85,978. Adjusted for population, Kenya has the densest developer community of the three, at roughly 11.9 developers per 1,000 people against 7.8 in Nigeria.
Accounts, however, say little about contribution. The more useful figures track commits. The same report found that contributors making at least one commit to open source repositories in Africa rose about 18% during 2025, from 9,286 to 10,955. Nigeria went from 2,049 to 2,371, Kenya from 862 to 1,022, and Rwanda from 71 to 112. Repositories with more than 1,000 stars rose from 127 to 177 across the continent, and Rwanda saw its first project cross that mark.
These numbers come from a dataset built around popular repositories, so they capture the visible end of activity rather than every commit. The gap between 9.3 million accounts and roughly 11,000 committers to widely followed projects is wide, although the two figures measure different things. Code pushes in Nigeria, Kenya and Rwanda have grown seven to eightfold since 2020, which suggests real development work rather than idle sign-ups.
AI is following the same path from a smaller base. Nigerian contributors to AI repositories rose from 137 to 180 during 2025. Nigeria produced its first AI repository with more than 1,000 stars in March 2025, and Kenya followed in September.
Communities do the groundwork
Much of this participation did not start with employers or universities. Open Source Community Africa (OSCA), co-founded in 2018 by Nigerian developer Samson Goddy, built a network of local chapters aimed at getting African developers, designers and writers to contribute to projects at home and abroad. Its OSCAFest in Lagos drew more than 1,000 people from five countries in 2020. By 2022, the festival had more than 1,400 members from nine countries across 54 chapters, according to the OpenUK report. OSCA’s GitHub organisation also hosts an open-source challenge and a curated list of tools built in Nigeria.
Outside programmes gave the culture a practical incentive. Google Summer of Code offered some Nigerian and Kenyan developers a paid route into upstream projects, and a 2021 month-long collaboration between OSCA and Meta led to several Nigerian contributors being approached by recruiters. In Kenya, groups such as Python Kenya and the Nairobi DevOps Community serve a similar function.
From contributors to infrastructure owners
The more consequential change is at the institutional level. Rwanda’s national digital payment system, eKash, runs on Mojaloop, an open-source platform for interoperable instant payments. Its operator, RSwitch, relaunched eKash in February 2025 and reported about 1.5 million transactions a month through September. RSwitch’s team also added its own modules for dispute management and reconciliation. One of its staff, Promesse Ishimwe, led the dispute management workstream inside the Mojaloop project, which means local work went back upstream rather than staying in a private fork.
Kenya has taken a policy route. Its Open Source Programme Office, launched in November 2025 within the ICT ministry, was one of two chosen from 28 applicants under a project funded by the European Commission and run by the ITU and UNDP. Officials say the government plans to move from roughly 22,000 digital services to 50,000 by 2027, and that licensing proprietary software at that scale would be prohibitively expensive. A survey of public-sector software use is meant to inform what they call a Kenya-preferred technology stack.
What is holding contribution back
The constraints are mostly economic. Open source depends on voluntary work, and the OpenUK report notes that Nigerian developers often drop projects once paid work arrives. Nigerian freelancers wait an average of 51 days for international payments, and currency volatility erodes what they eventually receive. Power cuts and unstable internet add to the cost of contributing.
Corporate support is thin. The report describes a near-total absence of domestic corporate investment in Nigeria’s open source ecosystem, leaving NGOs, volunteers and international sponsors to carry the load.
Institutional adoption also lags usage. In January 2026, downloads tracked by Scarf reached 16.9 million in Nigeria, 13.9 million in Kenya and 3.7 million in Rwanda. Yet only four Nigerian and two Rwandan public-sector organisations appeared in the data, and none from Kenya. Steve Haley of the Mojaloop Foundation points to procurement rules as a structural barrier, since many funders ask for prior implementation records that new local implementers cannot have.
What the numbers do and do not show
The evidence supports a measured conclusion. African contribution to open source is growing faster than the global average from a small base, and Nigeria, Kenya and Rwanda each show a different route: community-led, policy-led and payments-led. But visible contribution still involves a small share of the continent’s developers, and it depends heavily on volunteers and foreign sponsors. Whether that changes will depend less on developer numbers than on whether employers, funders and governments start paying for the work they already use.


