Why Zambia, Why Now (Part 1 of 4)
The infrastructure most economies take for granted doesn't exist here as one thing. That gap isn't just a problem — it's the opportunity.
The Thesis — a series on why we're building sovereign rails. Part 1 of 4.
Why Zambia, why now
The infrastructure most economies take for granted — a single way to prove who you are, move money, and satisfy a regulator — doesn't exist here as one thing. It exists as a dozen half-connected services, each with its own login, its own fees, and its own idea of the truth.
That gap is usually framed as a problem. I think it's the opportunity.
When you aren't carrying decades of legacy rails, you get to build the thing every mature market wishes it could rebuild from scratch: identity, money and compliance as a single sovereign network — owned locally, forensic-ready by default, and designed for the conditions we actually operate in rather than retrofitted from somewhere else's assumptions.
Why now, specifically? Two things moved at once.
First, the primitives got cheap and good. Postgres, Elixir on the BEAM, commodity cloud — you can stand up genuinely serious, concurrent, fault-tolerant infrastructure without a data-centre or a nine-figure budget.
Second, AI leverage changed who can build. A small number of skilled people can now credibly ship national-grade systems that used to demand large teams and budgets this market doesn't have. (I put hard numbers on that in the Delivery Efficiency series — it's the reason a project this size is even plausible solo.)
And "sovereign" is the load-bearing word. It means identity, data and settlement a country actually controls — not rented from a platform in another jurisdiction that can change its terms, raise its rates, or withdraw access on a schedule you don't set.
Honest note: this is a hard sell and a long game. Trust infrastructure is earned in years, not launches — and no amount of good engineering shortcuts the part where institutions decide to rely on you. I'm fine with that. The things worth building here were never going to be quick.
A thesis is just a claim until you write the first line of code. On 9 February 2026, I did.