Presentation

From French government hacker to founder, CTO, and blockchain engineer.

I was recruited by the French government at 20yo through international CTF competition and worked there full time while completing my engineering degree. Offensive security was the right starting point: it exposed me to a wide range of architectures and taught me to reason across the OSI layers, from high-level interfaces and application logic down through cryptography, networking, low-level protocol work, and hardware security.

That range made blockchain a natural place to work, because the engineering only really makes sense if you can think across protocol design, infrastructure, execution, and market structure at the same time. Today I focus on Ethereum and Solana distributed systems, especially cross-chain infrastructure, trading systems, and market design around prop AMMs and CLOB-style execution, where performance and cost optimisation are central. Right now that means working on opinion-market infrastructure and transaction-landing reliability for a company I work with, treating execution quality as a core product problem rather than an afterthought.

I like zero-to-one work. I have scaled engineering from 0 to 10 engineers, taken products and teams from idea to operation, led due diligence across multiple investors, and helped raise more than $13M from pre-seed through seed with a number of considered S-tier investors. Doing that well is less about any single trait than about combining product sense, engineering depth, speed, hiring, team alignment, and execution under constraint, while keeping maintainability, engineering quality, and security intact.

Rust and Go are the languages I use most, including work at the L1 and validator-client layers on both Ethereum and Solana. A lot of that work sits close to node operators, in critical environments where private-key handling, enclave design, failure domains, performance tuning, timing games, and revenue optimisation are first-order concerns. I have modified base clients to improve APY and operator returns, and introduced new features required by the ecosystems I was building in. Terraform, Terragrunt, and Ansible are part of my daily infrastructure work, and I am usually at my best on the complex technical problems that other teams would rather not own.

I work across French, English, and Spanish at C1, Danish at B2, and Korean in progress. Clear communication matters to me as much as technical depth: I am comfortable moving between engineers, operators, investors, and leadership teams, and framing the same problem at the right level for each of them.

Current mode

I do my best work when I can take a system end to end: research, specifications, implementation, infrastructure, and production delivery. The current focus is institutional-grade blockchain engineering where protocol constraints, execution quality, operator reality, and market structure all need to line up. Good systems have to work for users, builders, and node operators at the same time, which is why L1 roadmap awareness matters as much as the implementation detail.

Role Lead blockchain engineer, former Founder / CTO

Infrastructure AWS, GCP, Flux monorepos, Terraform, Terragrunt, Ansible, fast iteration loops

Application PropAMMs, orderbooks, contracts, CU optimisation, execution-critical product systems

Low-level Client performance, node communication, blockspace efficiency, low-level optimisation

Leadership Principal engineering, specs, hiring, political alignment, end-to-end delivery

Languages

MBTI INTP-AH

Ethereum Solana Execution Layer Coordination Layer Sequencer ZK Research L1 Performance Optimisation Security Enclave Prop AMMs CLOB Oracle

Careers

Chronology of work

This chronology makes more sense through the systems and constraints, not the industry names. I started by creating internet communities and organizing teams around websites and release workflows, before streaming platforms existed and when distribution had to be built from scratch. That was already entrepreneurial in practice: not just building the tooling, but also organizing the people, process, and delivery around it. I moved into offensive security because I wanted the kind of work that forces you to understand many different infrastructures and software stacks quickly, and because there is a particular challenge in breaking systems that are supposed to be bulletproof. From there I moved into distributed systems handling heavy computation, data aggregation, and real-time operational decisions, then into large travel infrastructure, platform security, and blockchain. Blockchain came later, but it felt like a natural continuation because it brings protocol design, infrastructure, performance, security, and execution into the same surface. One thing I keep seeing in this space is how little direct contact there still is between application builders and core devs, even though both sides shape the same system. A lot of my work in blockchain has been about closing that gap, moving across ecosystems and working both close to the base layer and at the application layer. The pattern across the whole path is simple: stay close to production, stay close to failure modes, and work on the parts of the stack where throughput, reliability, and operator behavior are directly affected.

Execution

Founder and hands-on CTO

Built a company from zero to 20 people and took engineering from zero to 10. I stayed hands-on through architecture, hiring, product definition, and delivery while also leading fundraising and investor due diligence through pre-seed and seed. The job was constant context-switching between company building and engineering without letting either side drift.

01 Company building Fundraising, hiring, architecture, and delivery while staying close to the engineering work.

Staking

Kiln: validator platforms and operator-grade systems

At Kiln I led backend engineering around validator infrastructure and validator-as-a-service systems across Ethereum and other proof-of-stake networks. The scope ran from shared Go and Rust foundations and internal CI to SLAs, monitoring, alerting, and the day-to-day boundary between software teams and operator reality. I was also already working on validator economics, improving APY and operator returns, and shipping targeted client or stack changes where there was room to improve.

02 2022 to 2023 / Kiln Validator platforms, shared foundations, operator tooling, and APY-sensitive performance work.

Protocol

Ethereum Foundation: Lighthouse and Beacon Chain internals

Through the Ethereum Protocol Fellowship, which I was doing while also working full time as a team lead at Kiln, I worked on Lighthouse and Beacon Chain internals in Rust, mostly around consensus-layer behaviour, performance, metrics, and debugging. That included open-source work on Lighthouse around the attestation simulator and performance instrumentation, alongside work on the implications of single-slot finality, especially what it could mean for hardware requirements, validator location, and the operator footprint of the chain. It was a good reminder that client code stops being theoretical once uptime, operator economics, and decentralisation constraints are involved.

StreamETH presentation Sigma Prime article: attestation simulator

03 2023 / Ethereum Protocol Fellowship Consensus-client internals, SSF impact work, and Lighthouse open-source contributions tied to operator outcomes.

Platform

Yinkozi: principal engineering for security products

At Yinkozi I led engineering and architecture for two security products across Go, Scala, React, gRPC/Protobuf, Terraform, Kubernetes, Prometheus, and ELK. In practice that meant shaping the architecture, building the platform, and keeping delivery standards high while the team was still moving quickly.

04 2019 to 2022 / Yinkozi Product architecture, platform delivery, and team leadership across a broad stack.

DevSecOps

Zendesk: security controls at infrastructure scale

At Zendesk I led DevSecOps work across EMEA, running threat modelling and security reviews with product and infrastructure teams. A lot of the work was about getting security controls into the actual delivery path: CI/CD, application architecture, key management, RBAC, and internal libraries. When the harder issues landed in infrastructure or code, I patched them myself instead of stopping at recommendations.

05 2017 to 2019 / Zendesk Security engineering embedded into CI/CD, infrastructure, and platform teams across EMEA.

Consulting

Octo / Amadeus: global slot auctions and distributed systems

Through Octo, I moved to Australia to work on Amadeus and a hard travel-infrastructure problem: the system used to bid, reallocate, and manage plane slots and parking capacity across airports at global scale. It had to ingest signals from everywhere, including operational feeds, weather, strikes, crew issues, and last-minute incidents, and turn them into decisions with real operational and commercial consequences. Alongside that, I worked on cross-region Kafka architecture, broker frameworks, Couchbase automation, AWS infrastructure, and performance testing. It was distributed systems work where latency, coordination, and reliability were part of the business model.

06 2016 to 2017 / Octo Global slot-auction infrastructure, wide signal ingestion, and distributed systems tied directly to revenue and operations.

Anti-abuse

Viadeo: real-time security and streaming systems

At Viadeo I built anti-abuse and security platforms across React, J2EE/CQRS services, real-time spam detection, expression-based rules, DynamoDB, and Spark and streaming workloads in AWS. That is where product security, data systems, and application architecture really started to collapse into one problem space for me.

07 2015 to 2016 / Viadeo Real-time anti-abuse, product security, and streaming systems inside the product surface.

Security

BULL / ATOS: pentest, incident response, security software

At BULL / ATOS I built tooling and deployment pipelines for pentest and incident-response teams while also working across vulnerability assessment, threat modelling, malware analysis, code review, and mentoring. The value of that period was the range: you had to understand how systems actually broke, across many different stacks, not just how they were documented.

08 2011 to 2015 / BULL-ATOS Offensive security, incident tooling, and software engineering across many stacks.

Teaching

External teaching: assembly and low-level security

At 24, while already working full time in security, I was also teaching as an external lecturer for final-year engineering students. The course was about assembly and low-level security, which meant teaching the mechanics underneath the abstractions most people use every day.

09 Age 24 / external teaching Assembly, low-level security, and practical teaching grounded in real work.

Government

Recruited through international CTFs

International CTFs led to me being recruited by the French government at 20, while I was still finishing my engineering degree. I was working full time in that environment and completing a security-focused engineering program in parallel. Starting there made rigor, discretion, and technical depth part of the baseline very early.

10 Age 20 / full-time government work + degree Full-time government work and a security engineering degree in parallel from age 20.

Adversarial

C, security tooling, and hacking competition

By 16 I was writing C, building security tooling, and experimenting with malware. By 18, while studying application and network engineering in a French IUT, I was competing internationally with CoP, Consortium of Pwners. That period built the low-level reflexes: exploitation, reverse engineering, systems thinking, and comfort with ambiguity.

11 Age 16 to 18 / IUT + CoP Low-level systems, exploitation, reverse engineering, and competitive pressure early on.

Start

Fansub teams, websites, and release ops

At 13 I built two manga and anime fansub teams, back when streaming platforms were not part of the picture and distribution had to be organized manually. For two years I ran the PHP and HTML sites, coordinated translation and adaptation, managed about 10 people, handled distribution, and learned release mechanics through Lua karaoke scripting, After Effects, and Python automation in IRC.

12 Age 13 / internet era Early product building, automation, and team coordination before any formal engineering path.

Projects

Selected technical projects

Public work that reflects the way I like to build: precise interfaces, operator-aware tooling, strong engineering surfaces, and systems designed around real constraints.

github-actionspec-rs Contract-style validation for GitHub Actions workflows and CI behaviour, with a bias toward making automation safer to reason about. knowledge-db-rs Rust-first knowledge tooling built around structure, retrieval ergonomics, and sharp interfaces instead of generic note sprawl. home-ops Bare-metal and lab operations, infrastructure automation, and node-oriented systems work carried out with the same discipline as production infrastructure. sol-validator-watcher Validator-aware monitoring and observability focused on the operating realities that matter to Solana node operators.