Engine Room

What we shipped since Banjul

In June we told you about Banjul: the first real certificates, issued with MBAviation for the staff of a West African airport operator, anchored on Avalanche mainnet. It was the proof that a certificate could live on-chain, in real conditions.

The summer served another purpose. It served to turn that demonstration into a product an organisation can join, use and leave cleanly, without us in the loop. Three pieces of work came out of it.

The complete certificate lifecycle. In Banjul we showed you that a correction is not erased, it is traced. That principle is now tooled end to end: versioned corrections, hardened revocation, full consultable history. An AeroCert certificate is no longer a frozen object, it is a living record whose every state carries its own proof.

Issuance governance. In a real organisation, an attestation is not issued by one person alone: it is validated, countersigned, sometimes corrected before it goes out. AeroCert now handles those circuits, from one-gesture signing to quorums anchored automatically once complete, including the replacement of an unavailable signatory, reminders, expiry alerts and a visible validation trail. Issuance follows the organisation's internal rules, not ours.

The governed template catalogue. Attestation templates aligned with aeronautical standards, bulk import by file, proof of the source document, and automatic flagging of uncertain fields at entry. This is the piece that makes it possible to move from a handful of demonstration certificates to real volumes without sacrificing data quality.

Around these three, the complete path: an organisation registers, is approved, and receives its workspace and its wallet in one click on the administration side. The reverse path exists too. Whoever leaves recovers all of their data, has thirty days, and then everything is purged. A trust infrastructure is judged on the cleanliness of its exits as much as anything else.

Who sets the rule, who checks it, who seals it

One question comes back every time we present AeroCert: what exactly are you certifying? The answer has three stages, and this summer's work mostly consisted of separating them cleanly.

At the start there is a rule, and it does not come from us. It comes from a prescriber: the party that defines what a course must contain, how long it lasts, what experience its trainer must have, and how long the attestation remains valid. On the courses we equip, that prescriber is MBAviation. As an approved competency centre, it is MBAviation that sets the rule for the training organisations tasked with delivering the regulated courses that companies in the sector need for their teams. Elsewhere it will be a sector authority, an airline, an airport, sometimes the training organisation itself.

Then the rule has to be checked. That is the role of the compliance engine: it compares what the prescriber requires against what actually happened for the learner. If everything matches, the attestation can go out. If not, it blocks. It is as simple as that, and it is precisely what was missing. Until now that check was done by hand, on spreadsheets, person by person.

Finally AeroCert seals. The proof is anchored, timestamped, and anyone can verify it with a scan, without calling us and without having to trust us.

This split is not a desk abstraction. It is what will let the same infrastructure serve an industrial or medical course tomorrow: the rule changes, the checking and proving mechanics do not.

The training platform takes shape

We are mentioning it in these letters for the first time: we are building a training platform for MBAviation. It handles the full cycle, from sessions to examinations and their marking, through to end-of-course attestations. Since this summer, every learner has an attestation wallet, managers have a view of their team's records, and anyone can publicly verify the proofs.

A word on how it is being built, because that says more than the feature list. A steering committee regularly brings together the people who do the work, trainers and operators included, and that is where scope and priorities are settled. We sit on it as a technical partner. A compliance platform built without the people who live compliance daily would stand no chance, and we would rather lose hours in meetings than ship a tool nobody can use.

Where it stands exactly, without dressing it up: it is still in testing, with MBAviation's operational teams reporting what does not work. It is the October-November pilot programme, with a first restricted circle of organisations, that will put it to the test of the real world.

The bridge with AeroCert is already there: every attestation coming out of it is meant to be a verifiable proof, of the same kind as those in Banjul.

Who does what, to be clear

A word of clarity, because the confusion keeps coming back. AeroCert is Kepler Aviation's timestamped proof layer, sold in euros to its clients. The training platform is MBAviation's, which commissions and operates it; Kepler Aviation builds it for them. AVIA is the token issued by Kepler Digitals, a separate company. Three different things, carried by different entities, each with its own accounts and its own obligations. This letter is Kepler Aviation's; when it talks about the token, it is to say what can honestly be said about it, not to promise anything.

The team

Kepler Aviation's team has grown this year. A further developer hire is under way, and apprenticeships will follow, on the development side as well as on commercial outreach.

What slipped, and what it changes

In April we wrote: "if we miss it, we will say so too." So let us say it, without dwelling on it.

The closed beta planned for August did not happen. The onboarding path had only just been finished, and opening up before it was solid would have served nobody. The first organisations come in during October-November.

Today, certificates are issued from a single surface: Clarity, the software we built for MBAviation. The training platform becomes the second one, and that is what the pilot opens.

The work on restoring the token's visibility on aggregators did not succeed, and we will not force it. In a market without volume, a restored listing would display a meaningless figure, and manufacturing activity to obtain it would be exactly the kind of practice we refuse. The energy went to the product, and we stand by that choice.

As for the decision on the token's role, it was to be taken on three months of production data. The clock started in June in Banjul. It will therefore be taken later, and always on data, never on speculation.

The market, without rose-tinted glasses

The crypto winter continues. AVAX is worth around 8 dollars today, against 9.6 dollars at the time of our April letter, and around 125 dollars at the 2021 peak. AVIA lives in the Avalanche ecosystem and is subject to its gravity, mechanically, regardless of what we ship on the product side.

And let us say it plainly: the AVIA market is today all but non-existent, with volumes close to zero. In those conditions no displayed price is meaningful, in either direction, and nobody should expect a market miracle in the short term. That does not change the order of things we set out: it confirms it. What we work on is the real use of the product. The rest is neither decreed nor promised.

Institutional blockchain, zero crypto marketing

One thing has to be said clearly, because it shapes everything we do: aviation does not buy crypto assets.

The players we work with, training organisations, operators, airlines, authorities, have a real and growing need for proof: who is trained, by whom, until when, with what validity. But those same players will never touch a token. Their purchasing rules exclude it, their accounting cannot handle it, their compliance refuses it, and their reputation forbids it. Both things are true at once, and that is not a contradiction: it is a specification.

The consequence is one we have owned since AeroCert's first day: the blockchain must be infrastructure, not a marketing argument. Our clients pay in euros, integrate through an interface, and have no wallet to manage, no crypto to buy and no smart contract to understand. The chain sits exactly where it is useful, under the bonnet, as a public proof registry anyone can verify with a scan, as in Banjul.

This model is not our invention. It is the model of the few players who have managed to get institutions to adopt a blockchain infrastructure: infrastructure communication, proof through partnerships and production, and never a word of crypto marketing. It is the only posture that works with institutions, and it is ours.

That is our line, and it is final: you will never see "to the moon" communication about this token. If the ecosystem's token were one day to have an operational role, that role would come after usage, never before, and it would be decided on data. It confers no rights today and it promises no service. This is not a renunciation, it is the only honest way to treat the subject.

The hypothesis of a pricing advantage tied to holding AVIA, raised in April, has not been retained. No such advantage exists today, none is promised, and were a decision ever to be taken, it would be taken on data and announced in writing.

A word on vocabulary, finally, because it matters. The people who hold AVIA are holders, not investors. The token confers no financial right over any company, and no return is promised.

The calendar set in April has not moved: Kepler Aviation's end-of-year decisions, production data and December commercial review, will define its 2027 action plan. A written update will be made on that basis, including AVIA's place in the ecosystem. No operation is decided or scheduled at this stage. No date pulled out of a hat, no promise detached from the product.

What comes next

Between now and the end of the year: the pilot programme launched in October-November with a first circle of organisations, the acceptance testing campaign carried through with MBAviation, and Kepler Aviation's December review put on the table, figures included, as promised in April.

The next letter will come out when there is something to tell. That has been the rule since the first one.