English 
한국어 
Français Paid to code at twelve.
I started by building Minecraft mods and server plugins in Java — and server owners paid for them. My first clients, my first users, my first production bugs.
- Backend
Case studies
Maxence Foulon
AI Product Engineer · Founder of Selvren
I design it, build it
and ship it.
I’m Maxence, a French builder. I sold my first code at 12, trained as a systems and network technician, designed for around ten freelance clients, and now run my own AI startup, Selvren. I take a product from the first user interview to the code that ships.
- 23 years old · Le Mans, France
- French native · English C1
- Remote today · moving to Korea
~ 10 +
freelance clients in two years
5 +
SaaS products launched
58 GiB
LLM run on a 48 GB Mac (MoEVM)
The journey
Every chapter added a layer.
No five-year degree. Each chapter taught me a new part of the stack.
-
-
How machines actually talk.
Vocational baccalaureate in Digital Systems (Bac Pro SN), with honours: electronics, hardware and networks — the layer most developers never see.
- Infra
-
Inside a university’s IT team.
Two-year Systems & Network Technician degree (EKOD, CCI Le Mans) as an apprentice in the IT department of Le Mans Université: support, a GLPI service desk, a workstation deployment server, network and server maintenance.
- Infra
- Data
Case studies
-
The fork in the road.
I started an engineering degree at CESI as an apprentice at Unixo, an IT & telecom integrator. Six months in, I chose building over diplomas — and Unixo kept me on a 4-month contract anyway.
- Infra
-
Designing for real people.
I launched my own freelance practice as a UI/UX designer — about ten clients in two years. At Wiim AI I was the only designer: the UX thinking and the entire app, solo, for six months.
- Research
- Interface
-
One person, a whole product team.
Mastering AI tools changed how I work. I architect the product, write the specs and run AI coding agents like a team — then review, test and ship. Solo builds like SouverainIA, team work with WebAPI, and 5+ SaaS launched.
- Backend
- Data
- Interface
Case studies
-
Answers you can trust.
Selvren is a B2B platform where teams ask AI domain experts grounded in their own private documents. Every answer comes with its sources — in a dashboard or through an API.
- AI
- Data
- Backend
Read the case studyCase studies
-
Down to the inference kernel.
Running Mixture-of-Experts models beyond the GPU’s memory budget on a 48 GB Mac: gpt-oss-120b (58 GiB) on the GPU, and Qwen3-Coder-Next 3.1× faster than llama.cpp’s op-offload path, with bit-identical output. A draft pull request is open on llama.cpp.
- AI
- Infra
Read the case studyCase studies
-
Where? — the whole stack, in one app.
An iOS-first Expo app that turns “let’s go out” into a real plan, alone or with friends. The map stays under clouds until you actually go there. 38 user flows and 78 edge cases designed before the first screen.
- Research
- Interface
- Mobile
- Backend
- Data
- Infra
- AI
How I build
I don’t just prompt. I run a team of AI agents.
The architecture, the specs and the final review are mine. The agents type fast; I make sure what ships is right.
- 01
Understand
Interviews, flows and the edge cases nobody wrote down.
- 02
Specify
Product rules, states and contracts — the source of truth.
- 03
Architect
Module boundaries and data model so agents can work in parallel.
- 04
Orchestrate
AI agents in isolated branches, each with a precise brief.
- 05
Verify
I review every diff. Types, lint at zero warnings, tests, smoke runs.
- 06
Ship
Docker, VPS, Cloudflare — with a written rollback for every deploy.
- 20 AI agents in parallel on Where?
- 1,749 tests on SouverainIA
- 0 lint warnings allowed
- 0.0 max deviation, MoEVM parity
Open source
Code you can run today.
Public on GitHub, MIT licensed, measured on real hardware.
MIT · New
watch-mcp
Lets Claude, Codex and Cursor watch your videos and screen recordings. With WatchRecorder, its macOS app: press ⌥⇧R, show the bug while you talk, your AI gets it.
- Built and shipped in one night
- 92 tests (66 Node · 26 Swift)
- ~10 s to watch a 2-min 1080p recording, transcript included
- 100% local: bundled ffmpeg, Whisper on Metal, no API key
- Works in Claude Code, Codex, Claude Desktop and Cursor
MIT
MoEVM
Expert paging for Mixture-of-Experts models on Apple Silicon, built on llama.cpp.
- 58 GiB model on a 48 GB Mac
- 3.1× faster than llama.cpp’s op-offload path
- Bit-identical logits
MIT
selvren-build-kit
TypeScript SDK, React components and a gallery to embed Selvren’s AI experts in any product.
- SDK + React components
- Open-sourced on 20 Sep 2026
The full stack
Hire one person, cover seven layers.
01
Research
Understand people before drawing a pixel.
- User interviews
- Personas
- User flows
- Edge cases
- Benchmarking (Mobbin)
- FigJam
02
Interface
Design it, then build it — same person, no hand-off loss.
- Figma
- Design systems
- React
- Vue
- Next.js
- Astro
- TanStack Start
- Motion
03
Mobile
iOS and Android apps with Expo, from first screen to store.
- Expo
- React Native
- iOS
- Android
- Reanimated
- Haptics
04
Backend
APIs, workers and business rules that hold under real traffic.
- TypeScript
- Node.js
- Fastify
- AdonisJS
- Python
- REST APIs
- Auth & 2FA
- Stripe
05
Data
Schemas, geodata and storage designed for what comes next.
- PostgreSQL
- PostGIS
- Redis / Valkey
- S3
- Migrations
- Data modelling
06
Infra
Trained as a systems & network technician. I deploy what I build.
- Linux
- Networks
- Docker
- VPS (Hetzner, Scaleway)
- nginx
- Cloudflare
- Rollbacks
07
AI
From grounded RAG products down to the inference kernel.
- RAG with citations
- Mistral
- Local LLMs (Ollama, MLX)
- llama.cpp
- Agent orchestration
- Claude Code
- Codex
Seoul
Remote today. Korea is the plan.
I live in France today and work remotely. I’ve been to South Korea three times, staying two to three months each time — and my plan is to settle in Korea.
- trips to Korea
- 3
- months per stay
- 2–3
Your afternoon is my morning.
A live overlap every working day — and the work keeps moving while Seoul sleeps.
Le Mans 9 AM
Seoul 4 PM
Seoul is 7 h ahead in summer, 8 h in winter.
Languages
- French Native
- English C1 · near-bilingual
Working with me
- Freelance
- Open to freelance projects: an AI tool designed, built and shipped, wired into the software your team already uses.
- Looking for
- An AI Product Engineer or full-stack role, in a team or a startup that needs someone who builds products end to end.
- Setup
- Remote from France, with regular 2–3-month stays in Korea. Settling in Korea is the plan.
- Visa
- No Korean work visa yet: working on site needs employer sponsorship (usually E‑7). Fully remote from France needs none.
- Start
- Available now.
Education
Learned by shipping.
- 2023 — 2024
CESI engineering track · 6 months
Left to build products full-time
- 2021 — 2023
Bac+2 · Systems & Network Technician
EKOD (CCI Le Mans Sarthe) · apprenticeship at Le Mans Université
- 2018 — 2021
Bac Pro · Digital Systems (SN), with honours
Lycée Saint-Joseph La Salle
Let’s build something.
I take freelance projects: I design, build and ship AI tools wired into the software your team already uses. Also open to joining a team or a startup that needs someone who builds — remote-first today, regular stays in Seoul, and a move to Korea ahead.
Or reach me directly
- Email maxence@selvren.com
- WhatsApp +33 6 19 32 62 26 Open ↗
- KakaoTalk ID Maxence.foulon
SouverainIA
AI executive assistant inside WhatsApp, hosted in France.
Context
SouverainIA’s founder wanted an AI executive assistant for SME leaders that lives where they already work — WhatsApp — with every byte hosted in France.
What I did
- Built the product solo: Python backend, WhatsApp (Meta) integration, a “Chef IA” that sets daily priorities and sends nightly summaries.
- Security: TOTP two-factor auth, session revocation, privacy-safe logs.
- Operations: Docker on a self-managed VPS behind nginx, versioned releases with a written rollback for each deploy.
- Designed and shipped the new landing page.
Result
Live at souverainiafr.com, covered by 1,749 automated tests.
Bastide
Private AI assistants for SMEs — offer, pricing and site.
Context
With a partner, a studio that builds private AI assistants for SMEs and firms handling confidential data — accountants first.
What I did
- Positioning, offer and pricing: paid audit, 6-week build, optional monthly follow-up.
- Brand and the full site: Next.js static export, industry pages, a “Bastide vs public AI” comparison, a security page.
- An SEO + AI-search content loop: scheduled drafts, competitor watch and a monthly visibility report — nothing publishes without human review.
ESLITECH
Manufacturer site redesign on their live CMS.
Context
ESLITECH (brand Smartfurry), an aquarium and pet-products manufacturer, had a dated site. The brief: same content and pages, a premium look.
What I did
- Rebuilt the front end in Vue 3 + Vite on their live CMS API, keeping every URL, call-to-action and popup working.
- A motion system, bento product sections and a dedicated About page.
- Shipped a public demo on Cloudflare Workers: HTTPS, noindex, forms disabled.
Selvren
AI experts on private documents, every claim checked against its sources.
Context
Teams want answers from their own documents — with proof. Selvren started as a chatbot for pet shops, became AquaChat for aquarium stores, then a B2B platform of AI domain experts.
What I did
- Founded the company and led the build with a team of AI coding agents: I set the architecture, the rules and the merge gate, and reviewed the work.
- Hybrid retrieval over private documents and an expert catalogue, then a rule-based verifier that checks every claim against its sources — it abstains or quotes when sources are thin.
- Security by design: row-level security in PostgreSQL, encrypted originals, antivirus on uploads, an egress proxy, hashed and revocable API keys.
- GDPR tooling: data export, erasure, a written retention doctrine.
Result
3,182 commits in 11 weeks, 11,901 tests in the merge gate, 28 architecture decision records.
Atelier
Raw idea → product brief through 8 specialist engines, 100% local.
Context
Every product I build starts as a raw idea. Atelier turns it into a brief an engineering team — human or AI — can execute.
What I did
- A local-first chain of 8 specialist engines, from discovery to product, with every step’s data persisted.
- Runs entirely on a MacBook (M5 Pro) with Ollama and MLX, under a strict zero-cost rule: no paid API, no rented compute.
- Benchmarked local coding models to pick the engine (qwen3-coder 30B: 5× faster than qwen2.5-coder 32B).
Result
Idea → product brief end to end in about five minutes, fully offline.
Naiad
Connected-aquarium companion app with an AI assistant.
Context
Naiad is a companion app for a hardware partner’s connected aquariums — built as a partner demo.
What I did
- Designed the whole flow: account-first onboarding with EU data residency, product recognition by QR code, tank profile.
- A living dashboard (water parameters, a confidence score), species compatibility, device control for lights and pumps, a logbook.
- An AI assistant that explains trends and prepares safe actions — but never controls a device without the owner’s confirmation.
GLPI
Set up GLPI for the service desk and IT inventory.
Context
The IT department supports staff across a whole university campus — every request and every piece of equipment has to be tracked.
What I did
- Put GLPI in place: a ticketing service desk for user requests and an inventory of the IT equipment.
- Used it day to day with the team for support follow-up.
PC deployment
A deployment server to install staff workstations.
Context
New and reinstalled workstations had to reach staff ready to use, the same way every time.
What I did
- Set up a deployment server to install work PCs in a repeatable way.
- Prepared and installed staff workstations.
Support & maintenance
User support, network and server maintenance.
Context
Two years as an apprentice inside the IT team of a university, alongside a Bac+2 in systems and networks.
What I did
- Day-to-day support for university staff.
- Maintenance of the network and the servers with the team.
- Hardware maintenance of workstations.
DataRace
A 223-mod “Create Civilisations” server, tuned and hosted.
Context
Still building for Minecraft: a modded survival server built on top of the “All of Create” pack.
What I did
- Assembled and shipped seven versions in five days (June 2026): colonies, world generation, trains, far-view rendering, a performance pass, then content and quality-of-life mods.
- Wrote down every choice: no land-claim mods, hard mob packs blocked, each mod tested on a staging world before going public.
- Tuned and hosted it: server memory 4 → 10 GB with GC settings, view distance 8, simulation distance 6, on a VPS with a game-server panel.
Result
223 server mods on Minecraft 1.21.1 (NeoForge), measured at 20 TPS (5.3 ms per tick).
Pokémon League
A competitive league mod in Kotlin: gyms, ELO, tournaments.
Context
A full competitive mode on top of Cobblemon, the Pokémon mod for Minecraft.
What I did
- Started the project with another developer and built its economy system.
- The mod covers a 7-stage league with 5 gyms, ranked 1v1 and 2v2 with ELO, double-elimination tournaments for up to 32 players and 96 quests.
SyncCast
Keeps esports casters on the same frame across Twitch and YouTube.
Context
Esports casters commentate the same match from different places — and each platform delivers it with its own delay.
What I did
- Built the product in 8 days (268 commits, December 2025): rooms, a leader caster as reference, and resyncs scheduled on a shared server clock (Ably).
- Twitch drift measured every second; a spread over 500 ms triggers a scheduled realignment.
- An own HLS relay: streamlink + FFmpeg into 1-second segments on Cloudflare R2, segments uploaded before the playlist.
- Teams with Admin / Manager / Caster roles and invites; room chat and live drawing over the video; Row Level Security on every table.
- Redesigned the landing page in 2026 to explain the backend — only with what the code actually does.
Result
Beta, with 52 unit and 14 end-to-end tests.