TransFund: crowdfunding on the Internet Computer

Context

In May 2024 I attended training on the Internet Computer (ICP), a blockchain platform that can host full applications, not only transactions. Within a few months I had entered two hackathons built on it.

The first, on 28 June 2024, produced an estate planning platform: elderly parents make their wills, and their heirs can follow them later. A will must not be changed after the fact, and storing it on a blockchain makes it tamper-evident. We took part remotely because we had classes at the same time, and finished third. Both competitions were built on the same platform, so what we learned in June carried into September.

What we built

On 27 September 2024 we won the ICP Rwanda Hackathon with TransFund, a crowdfunding platform built on the blockchain.

Crowdfunding runs on trust. People who give money want to know that it reaches the cause and is used as promised. Recording each contribution on a blockchain makes the record transparent and hard to tamper with. That record shows what was given; it does not on its own show how the money was spent, which is a separate problem.

In plain terms, a blockchain is a shared record that no single party can quietly change after the fact. For crowdfunding, that means a donor, a campaign owner and an observer can all check the same list of contributions and see the same thing.

I led the team. I worked with UWAYEZU Jean Felix, TURIKUMWENIMANA Daniel and NSHUTI Emmanuel, and we built it together within the hackathon.

The hard part

Most people who would fund a cause in Rwanda do not hold cryptocurrency. They pay with mobile money or cards. A blockchain product that only accepts crypto would exclude most of the people it is meant for.

So the part I am proudest of is the bridge between off-chain payment providers and the on-chain platform. A contributor pays the way they normally do, and the contribution is recorded on the blockchain. That connection is what turns a blockchain demonstration into something an ordinary person can use.

Results

• 27 September 2024: first place at the ICP Rwanda Hackathon with TransFund, with 500,000 Frw.
• 28 June 2024: third place and 200,000 Frw at an ICP hackathon, taken remotely, with the estate planning platform built by the same team.
• Three months separated the two results. The third place in June was feedback, and we came back in September and won.
• A working crowdfunding platform that connects familiar payment methods to on-chain contribution records.

What I would do differently

I would design the payment bridge before the on-chain features, because it decided whether ordinary people could use TransFund at all. The blockchain part is the novel part, but the bridge is what makes it usable.

I would also carry the lessons from June's third place into the first day of the next competition. We came back three months later and won, and the feedback from the first one is a large part of why.

Stack and role

• Role: team lead.
• Team: UWAYEZU Jean Felix, TURIKUMWENIMANA Daniel and NSHUTI Emmanuel, with me as team lead.
• Stack: Internet Computer (ICP), blockchain, and mobile money and card payments connected to the on-chain platform.
• Timeframe: May to September 2024.