Who We Are
Euler Advisory is an independent quantitative risk practice for digital assets and traditional finance. It brings to digital assets the way supervised finance builds, tests and governs its models.
The founder
I founded Euler Advisory because I have worked on both sides of the line between traditional finance and digital assets. At the ESMA-supervised credit and ESG rating agency where I spent over four years, first as a quantitative researcher, then as Head of Model Validation Europe, no methodology was used until someone outside the team that built it had challenged it, in writing. In digital assets, I see tokens raise money and reach exchanges on economics that nobody independent has tested. The firm carries the first discipline into the second market.
My doctorate was in game theory for decentralised systems; my postdoctoral research, at CentraleSupélec and RTE, modelled the French power grid. Game theory is how I read the incentives in a token design.
- Rating agencyHead of Model Validation Europe (Feb 2022 – Jun 2025), after a year as quantitative researcher (Mar 2021 – Feb 2022), at an ESMA-supervised credit and ESG rating agency; dealt directly with ESMA in its supervision of the agency.
- Postdoctoral researchApplied mathematics and AI at CentraleSupélec and RTE (2019–2021): partially observed dynamical systems and modelling of the French power grid.
- DoctoratePhD in applied mathematics, ENS and CNRS, in the ERC-funded project EQUALIS (2016–2019); game theory in decentralised systems.
- Master'sMathematical modelling, Sorbonne Université and École Polytechnique (2015), with an FSMP scholarship.
- Quantitative analystStratagemma Group, Santiago, Chile (Jun 2012 – Dec 2013): time-series modelling of NYSE and NASDAQ data, backtesting of algorithmic trading strategies and execution-cost optimisation.
- TeachingWorkshop instructor, 'Can Crypto Be Modelled?', ETH Zürich FinsureTech Hub, May 2026; supervisor of four student research groups, ESILV Finance department (2026–27).
- ResearchPapers since 2017 in applied mathematics, finance and digital assets, including articles in the Journal of Computational Finance and Nonlinear Analysis: Hybrid Systems, a Cambridge University Press book chapter, and the doctoral thesis; research in progress.
- Based inParis. Works remotely with clients in the EU and other markets, in French, English and Spanish.
What we believe
A model's limits belong in front of the people who rely on it.
Before it moves money, it is challenged by someone who did not build it, and what it leaves out is written down, in digital assets as in traditional finance.
-
A compliant structure can still fail.
A structure can meet every rule and still break under stress. A review tests both. → The 2025 case file
-
Uncertainty is sized, not removed.
We state what the model leaves out, and how far the result moves when an assumption changes.
-
Errors are cheapest before the decision.
Before the decision, an error is a change to a model. After it, it is a supervisory finding, a failed redemption, a mark-down or a lost round. The question is who finds it first: you, us or the market.
-
Evidence, not persuasion.
An allocator's risk team discounts what a deck says and checks what the model supports. We show which claims hold; we don't write them.
How we work
One side of the table
We act for one party in any transaction. On any one structure, we test it or we build it, never both. A conflict check is documented before work starts.
Signed by name
Every finding is checked, and every report is signed by its author, who answers for it to your board, your supervisor or your counterparty. AI tools make the work faster; they never replace the review.
Conservative numbers
Fat-tailed distributions, explicit model limits, and every assumption written where a reader can challenge it. When the data can't support a number, we say so.
Stated limits
Every deliverable states what was tested, what held, what didn't, and what we couldn't test. Each finding says what happens if it is left as it is.
What we are not
- A consultancy that bills juniors. The person you meet does the work and signs it.
- A code auditor. We test the economics, not the smart contracts.
- A rating or data vendor. We publish no scores.
- An investment adviser. We assess risk; we never recommend buying, selling or allocating.
Contact
Start with a call
A 30-minute call.
Confidential: we sign your NDA before any data changes hands.
Prefer email?
contact@euleradvisory.eu