A licensed payment institution built from nothing
A complete payment company built for a central bank licence: the settlement core, merchant and agent back offices, and client apps for iOS, Android and Telegram.
Task, solution, architecture and a result with a number, for every project.
No product name and no site address appears in any of these case studies. Some of the work was commissioned and we do not publish client names. Some of it belongs to payment projects that prefer quiet. The task, the architecture, the stack and the result stay where they are, and that is the part a technical buyer reads.
More than half of these projects are no longer running, and the reason is not technical. They were our own products, and they ran out of money for promotion rather than out of engineering. A client buys the ability to build a system, not somebody else's marketing, which is why closed projects sit here alongside live ones.
Range
The run does not break for a single year from 2014 onwards. Each period below is named after what showed up in it for the first time.
A handicraft store taking international cards and shipping abroad, a store for digital goods, corporate sites for a training centre and an investment firm, and the first messenger bots.
Billing for an e-wallet, an online identity verification service, foreign services paid with local cards, an aggregator pulling data from many independent sources, and a delivery map for a manufacturer.
Flight tickets paid with local cards, a network of self-service terminals accepting banknotes, delivery and taxi bots, online stores with stock control, and a prototype for ordering bank cards online.
A payment institution licensed by the central bank, a merchant platform with dozens of systems connected, a subscription platform for an expert, and corporate sites for manufacturers.
A student forum split by region, university and faculty. A program for the chip inside a wastewater sensor, which is not web work at all but hardware. A store for an occupational safety company.
A classified ads aggregator with a language model in the processing pipeline, an adaptive testing SaaS, a five-language content platform, a CRM for online sales, and our own knowledge base with a team of AI specialists in it.
A complete payment company built for a central bank licence: the settlement core, merchant and agent back offices, and client apps for iOS, Android and Telegram.
Kiosk software that takes banknotes, clears payments and defends itself. Fifty terminals in Andijan ran on one billing system.
People obtained a verified wallet status through a website: apply, pay, submit documents, get a decision. At its peak the service handled up to a thousand applications a month.
A shop integrates once and reaches dozens of payment systems. Fees, merchant balances and scheduled payouts are all calculated inside the platform.
Leads are split into queues, an order is locked to the operator working it, and closed sales are sent back to the ad account. The advertising learns from real deals.
A place where a student finds the discussion of their own faculty at their own university instead of a ten-thousand-member group chat.
Search across every destination and a ticket bought in the national currency with a local card, without conversion and without a foreign bank.
An aggregator that collects offers from dozens of independent sources, brings them to a common shape and shows the comparison in one table.
A service that let people pay for foreign online services with Uzcard and Humo when they had no international card.
Pottery and craft work from Uzbekistan sold to buyers overseas: international card payments, cross-border shipping, and the earliest project in this portfolio.
An expert publishes their calls, and a subscriber pays whichever way suits them: per item, per month, or as a share of the result.
The twenty case studies above are the projects with a technical story behind them. Here is the rest of what we can still name, one line each, newest first. Client names and site addresses are left out for the same reason as in the case studies.
Tell us about the task. We will study it and propose an option that makes sense technically and economically.
We reply within one business day · Telegram @sbunyod