AI agents, developer tooling, and the systems underneath them.
Montréal · Available from January 2027
About
I’m Eliot, a software engineer focused on AI agents and developer tooling. I like long bike rides, short feedback loops, and knowing why things work.
Most recently I spent six months at Botpress in Montréal, where I got to both build with the product and build the product. I shipped internal AI agents that run in production, built debugging tools for the agent SDK, contributed to the open-source LLM engine, and helped launch Viber, Botpress’s new agent builder you drive in plain English. Best part: shipping something and watching people use it an hour later.
I first came to Montréal for an exchange semester at Polytechnique, during my master’s-level engineering degree at Grenoble INP – ESISAR. I stayed for Botpress, and I’d like to make it home. French is my first language and English my working one. I’m also eligible for a Young Professionals work permit, so hiring me doesn’t need an LMIA.
Before that, I worked on payments at Upflow, a YC-backed fintech in Paris, and replaced Koesio’s spreadsheets with a printer configurator and margin-calculation platform: different stacks, different cities, enough variety to feel at home in the unfamiliar.
I care about doing things right: not perfectly, just properly. I can’t pretend something is fine when I know it isn’t, which makes me a bad salesperson and, depending on who you ask, a good teammate. And when I want to understand something, I build a small, worse version of it: an HTTP server from scratch in C, a text editor in Rust, my own fine-tuned LLM.
In October 2026 I rode from Grenoble to Sevilla: 18,500 m of climbing, 136 hours in the saddle, a hammock and a tarp for a home, and a hostel every time the rain won. Main takeaways: buy a lighter hammock and a bigger power bank.
When I’m not at a keyboard, you’ll find me:
shooting filmbuilding a new FPV drone (and crashing it)trying to learn guitarfinally getting better at climbingout on a bike rideplanning a bike route that's slightly too longhiking whenever mountains are within reachchasing a sub-39 10k, a 1h36 half, and maybe finally signing up for a marathonskiing whenever conditions allowat Piknic Électronik on a Sundaycooking something new (or trying to)hunting down Montréal's best barslost in some rabbit hole on the internetexploring Montréal's music scenetrying every ice skating rink in Montréalcamping somewhere with no signal
Always up for talking code, travel tips, Montréal gems, or whatever rabbit hole I'm in this week. Feel free to reach out.
Software Engineering InternMar – Aug 2026Montréal, Canada
Built Midas, a production enrichment agent running daily across 1,200 contacts. LLM classification, workspace resolution, and CRM write-back replaced a manual sales-ops process.
Shipped Drag, a company-wide RAG security agent that answers policy questions with cited sources and auto-fills vendor security questionnaires, saving 19h/month for legal and people ops. Its self-healing loop routes unanswered queries to experts on Slack and feeds their replies back into the knowledge base.
Contributed fallback model routing to Botpress's open-source LLM engine (15k stars), letting users define provider priority chains that auto-retry on failure.
Software Engineer InternJan – Jun 2025Valence, France
Built a benchmarking platform (Vue.js / Laravel / PostgreSQL) that replaced manual spreadsheets with an auditable system, powering group margin calculations on a $300M project reviewed by executive and project committees.
Authored the software specs and led a 3-person Agile team through sprint planning and task dispatch.
Fullstack Engineer InternJun – Aug 2024Paris, France
Built a net-new payout tracking view (React / GraphQL / Express) handling $1M+ in monthly transactions, giving customers their first centralized view of those flows.
Designed and integrated a payment refund system on the Stripe API, removing a manual support step.
Freelance
Fullstack DeveloperOct 2022 – Dec 2023Remote
Built and deployed client websites with Astro, TypeScript, and TailwindCSS, tuned for SEO, accessibility, and performance.
Migrated an ageing WordPress site for an overseas client and added search, lifting organic traffic by 25%.
An end-to-end ML project: I fine-tuned small, open-weight LLMs to turn a git diff into a clean Conventional Commit message — running fully offline as a git hook, so your code never leaves your machine. It spans the whole pipeline: ~12k diff/message pairs scraped from 13 disciplined TypeScript repos, LoRA fine-tuning with Unsloth, a three-layer evaluation (ROUGE-L, structural checks, and an LLM-as-judge), and deployment with Ollama. The headline result: a tiny fine-tuned model beats prompting one several times larger.
An information-security assistant I designed and shipped on my own at Botpress. It answers employees' policy questions in Slack and fills in customer security questionnaires, citing the source policy for every answer. When it can't fully answer, a self-healing loop sends the gap to the security experts and turns their reply into the next person's answer. It saves about 19 hours of work a month.
A playable Othello game with an AI opponent. The game logic is in TypeScript and the board UI in SolidJS. The AI is a minimax search written in C and compiled to WebAssembly. It runs in a Web Worker, so the page stays responsive while it looks moves ahead. Three difficulty levels: pick one and try to finish with more discs than it does.