Houssem Ben Salem, MSc

Lumin is a systematic directional model for cryptocurrency perpetual futures, built on a multi-agent LLM cohort architecture with a structured reasoning curriculum and consensus-driven capital decisions. Live production on BTC, ETH, SOL since January 2025. Validated across a 6.5-year chained backtest spanning the COVID crash, the 2022 FTX/LUNA capitulation, and the 2024–26 cycle. The path to this architecture began in the same craft I had already been practising for years.

Reading the physics behind a metrology measurement uses the same logic as reading the psychology behind a market price: identify the generator, model its influence, engineer for it. The reading reflexes, modelling tools, and post-mortem discipline I built across 14 years of Machine Vision engineering have been directly reusable in Lumin. Since 2021 the two trajectories have advanced in parallel, each breakthrough on one side feeding the other.

On the Lumin side, the path passed through three architectures — each built on the last, none discarded. The first was Python trend detection: arithmetic that decomposed price action into structural primitives — bricks, ratios, range regimes, momentum. The second composed those primitives into deterministic strategies, then decomposed the strategies into their driving components and mined the recurring configurations as signatures — statistical multi-bar patterns validated out-of-sample across four timeframes on Binance perpetual futures. The third and current architecture is the multi-agent reasoning model: LLM specialists that read the market bar by bar, integrating components in context — with the signatures as their language and the curriculum as the library they read from. Every prior phase is preserved — externalised in a wiki the LLM consumes natively.

What lives in the wiki is, in effect, a practitioner's mind made portable. The approach generalises anywhere skilled experts produce value. The white paper has the architecture in full.

2015 — present

At SYLVAC SA — Swiss precision instruments, Yverdon — I joined in 2015 as an applied optics engineer building vision metrology systems. In 2018 I moved to project management and delivered a generation of optical measurement machines end-to-end: specifications, industrialisation, production. Since 2021 I lead the R&D machines team — still building alongside, pushing the domain into embedded agents, adaptive workflows, machines that interact with humans more naturally.

Watch the testimonial →

Machine vision · optical metrology · quality control · robotic guidance · automation

2012 — 2015

Before SYLVAC, I was an applied optics engineer at Hexagon Manufacturing Intelligence. I worked on metrological machines, optical sensor development, and image processing for precision measurement systems.

2022 — 2024

Applied Data Science: Machine Learning program at the EPFL Extension School — the foundation for the data and Machine Learning work that became Lumin.

2010 — 2012

MSc in Microengineering — Photonics and Applied Optics at École Polytechnique Fédérale de Lausanne (EPFL). Photonics (light guides and lasers), advanced image processing, optical system design.

2007 — 2010

BSc in Microengineering at École Polytechnique Fédérale de Lausanne (EPFL). Optics, vision systems, mechanics, electronics, software engineering.

I work at the intersection of quantitative systems, AI architectures, and precision engineering. My strength is not staying inside one domain but connecting several until they become one solution. I'm an engineer first: every problem reduces to understanding the physics, building the math, writing the code, and delivering it personally — and staying on it longer than most people think is reasonable. I think outside the box, inside the path. Multilingual. I used to code in C++ and MATLAB. Now fully in Python, mostly through AI. English is becoming the new language of code.