About
I'm a software engineer and researcher in Bordeaux. I fell into computer science young and never left, and I've spent the years since on a single question: how do you establish trust without a central authority? A wallet refuses a request it cannot verify. An AI engine will describe a company to a stranger with no verification at all. Unverified actors and unverified statements are the same problem seen from opposite ends, and I work on both.
The first end is Verana, at 2060.io, an open trust layer for the internet. This year I wired Verana trust resolution into third-party identity wallets and solutions, among them the EUDI reference wallet, MOSIP's Inji, Talao, Sphereon, Paradym and Procivis One. Every one of them shipped with a hardcoded list of issuers it trusts. Replacing that list with a live, fail-closed call is the whole idea. I co-authored the Verana trust specifications, built the wallet conformance suite that checks those integrations hold under real flows, took Verana into France Identité's EUDI playground, and contribute to credo-ts at the OpenWallet Foundation, part of LF Decentralized Trust since September 2026. I'm a member of the W3C Credentials Community Group, the Trust over IP Foundation, the Agentic AI Foundation and the Linux Foundation.
The same problem is arriving for AI agents, faster. I co-authored a protocol showing that two previously unknown agents can establish mutual cryptographic trust in two round trips, with under 2ms of overhead. The other end is CiteMe, which I co-founded and where I'm CIO, measuring what AI engines say about a company. That answer becomes a reputation for whoever reads it, nobody at the company was asked, and nothing on the open web lets a model check who is behind a claim. I'm 20, and I've bet everything on this being one of the foundational problems of the next era, for people and for the agents acting on their behalf.
Before tech, I trained as a competitive parkour athlete with the French Gymnastics Federation from 2017 to 2025. In 2025 I co-founded PKBA, now the largest parkour club in southern France. Discipline, pattern recognition, and learning from failure fast: sport and systems thinking have more in common than they appear.
Research Interests
- • Trust Registries & the Trust Registry Query Protocol
- • Cross-Ecosystem Wallet Interoperability
- • Agent Identity, Delegation & Scoped Authorization
- • Credential Exchange Protocols (OID4VP, OID4VCI, SD-JWT VC)
- • Trust Frameworks & Governance Models
News
Latest posts
Seven Out of Seven, Then One in a Hundred
For five days in October, ChatGPT's links stopped saying where they came from. I built a lab to identify AI assistants from their network calls, and it worked every time. On real sites, it worked about once in a hundred.
Nobody Knows How Paracetamol Works
We trust drugs we do not fully understand because they passed a trial. I think software is heading the same way. On verifying instead of explaining, why Terence Tao is right to worry, and what you still owe.
The Job Moved Up One Floor
I can tell you exactly what my own product does. I could not tell you how it is built. On reviewing instead of writing, what that actually costs, and why I think it still works out.