We spent a few days at the Global Blockchain Conference in Istanbul, mostly to calibrate. It is easy to read the space from a distance and conclude either that everything has changed or that nothing has. Neither is true, and the gap between the two is where the useful work sits.
The short version: the infrastructure layer has quietly become boring, and that is the most encouraging thing about it.
Settlement is becoming plumbing
The conversations that felt most grounded were not about tokens at all. They were about settlement — moving value between parties who do not share a ledger, with finality and an audit trail. Framed that way it stops being ideological and starts being an integration problem, which is a much healthier place for a technology to be.
The teams furthest along had stopped treating the chain as the product. It was one component behind a normal API, with normal monitoring, and the interesting engineering was in the off-chain indexing, the reconciliation and the operational tooling around it.
The demos are still ahead of the engineering
The other half of the room was the familiar pattern: a compelling demo, a roadmap, and no answer to what happens when a transaction fails halfway, or how the thing is audited, or who pays the gas at scale. Those are not edge cases. They are the majority of the build.
Our rough filter has not changed. If a project cannot explain why it needs a distributed ledger rather than a database with good access control, the honest answer is usually that it does not.
The useful question is never whether something can be put on-chain. It is what breaks if you do, and who carries that cost.
What we brought home
- Treat the chain as one system among several. The hard parts — indexing, reconciliation, monitoring, key management — are ordinary distributed-systems engineering.
- Budget for audit from the start. Contracts written to be audited cost meaningfully less than contracts retrofitted for it.
- Design the failure paths first. Partial settlement, reorgs and stuck transactions are normal operating conditions, not exceptions.
- Be honest about whether you need it. Most workflows that get proposed for a ledger are better served by a database and a clear permission model.
We came back more interested in the space than we went, but for unglamorous reasons. The parts that are maturing are the parts that look like infrastructure engineering — and that is work we already know how to do well.
If you are weighing up whether a ledger belongs in your architecture, that is exactly the kind of question our AI and blockchain engineering work starts with. You can also see how we scope this kind of project, or get in touch.