An ecosystem full of African builders that the rest of the world could not see.

The arrangement

I contributed to Starknet through the Wolf Pack League, Starknet's contributor programme, not as an employee of Starkware. Worth saying plainly, because it shapes what the work actually was: no brief, no editorial calendar handed down, nobody assigning topics.

What got made, got made because I decided it should exist.

The problem: an active ecosystem that looked inactive

Starknet is a ZK-rollup Layer-2 with real builder activity across Nigeria. From inside the global ecosystem, you would never have known it.

Three things were happening at once. Products were launching regularly with no onboarding content aimed at African users, so the region stayed permanently behind the adoption curve. Global ecosystem updates were written at a technical register that assumed prior context. And African builders had almost no visibility in the wider Starknet conversation, which quietly created the impression that nothing was being built here.

That reframe mattered, because the two need opposite solutions. A participation problem needs recruitment. A documentation problem needs someone to write down what is already happening.

The strategy: two standing series, two different jobs

Rather than publish reactively around whatever launched that week, I built two recurring formats with distinct assignments.

Inside StarknetAfrica

Pointed inward at the region and outward to the world. Each edition profiled a specific African product, team or builder: what they were making, why it mattered, and a direct link to try or follow it. The job was visibility: give African builders a surface in the global conversation, and give global readers evidence the activity was real.

Stark Week

Pointed outward at the ecosystem and inward to the region. Weekly updates on integrations, new projects, governance and milestones, deliberately written so a beginner and an experienced builder could read the same piece without either being lost or bored.

One made the region legible to the ecosystem. The other made the ecosystem legible to the region. Together they closed the loop.

Product showcase, led by video

Recurring series build the audience. Product content converts it. When something shipped on Starknet I made the piece nobody else was making for this audience, and for this work I led with video rather than writing.

That was a deliberate call, not a preference. Onboarding failure is almost never conceptual. People do not abandon a wallet because they misunderstand custody, they abandon it at the screen where they cannot tell which button is the right one. Text describes an interface. Video shows it, at the speed a first-time user actually moves.

Fourteen pieces, nine products, each one followable with your own wallet open:

• Ekubo. Limit orders, so you buy and sell at your price instead of whatever the market hands you
• Endur. Liquid staking STRK for xSTRK rewards
• Argent. Gifting crypto in 34 seconds, and a first reaction to the Metal card
• On-ramp. Buying STRK with a card, no bank transfer needed
• Brother ID. Getting one, step by step
• PulsarMoney. Sending creators a coffee, on-chain
• Mancala. A live gameplay deep-dive with the builders

Alongside those, the concepts the products assume you already hold: token-bound accounts, session keys, why self-custody is not optional, and whether your wallet is actually protected.

See all fourteen →

Showing up in the room

Content about a community written from outside it reads like content written from outside it. I attended and covered Starknet builder events across Nigeria: workshops, builder circles and meetups, running mini-interviews with the people who turned up.

• Starknet Lagos Meetup 3.0. Volunteered on the event and hosted a session on games built on Starknet
• Starknet Builder Circle, Abuja. Spoke on BTCFi education
• Starknet Builders Workshop. Covered the event and interviewed attendees

That is where the source material came from. The events were not marketing for the content; they were the reporting.

Results

Inside StarknetAfrica became a standing asset rather than a campaign. The kind of thing that keeps returning value long after the publish date, because it is the only written record of a specific moment in that ecosystem.

What this work taught me

Localisation is not translation. Nobody needed Starknet's docs in another language. They needed content that started from a different set of assumptions about what the reader already had: which on-ramp they could actually use, which products were reachable from where they were, whether anyone like them was already building.

I also learned to pick the format from the failure point rather than from habit. If someone gets stuck understanding an idea, write. If they get stuck operating an interface, show them. Most onboarding content gets this backwards and then wonders why the funnel leaks.

And the pieces that travelled always had a real person in them. I only had those because I turned up to the meetups.