About Jonathan

I have always learned by building.

I am a quant, teacher, dad, Brazilian jiu-jitsu blue belt, and musician. I like taking hard ideas apart, building something with them, and finding the explanation that makes another person lean in instead of politely nodding.

Jonathan Laliberte-Alle — Different arenas, same habit: understand the system, try something, and learn from what happens.

The short version

Maths came first. Teaching made it personal.

I grew up in France and was drawn early to mathematics and computer science. I liked that a small set of ideas could explain a complicated system. Code could turn those ideas into something you could test.

That interest led to engineering school, graduate work in probability and finance, and a career in electronic markets. For years, my job has involved making decisions in noisy environments: deciding what can be measured, which assumptions are safe, how a system should react, and what to do when the model meets a case it has not seen before.

Parenthood changed the questions I cared about. I became less interested in whether a child could repeat a procedure and more interested in whether she could inspect a claim, choose a method, recover from a mistake, and explain her decision. I also became determined to help my children treat mistakes as part of learning rather than something to avoid. Teaching at Primoris Academy gives me a place to work on those questions with students from grades 5 through 12.

What I studied

Mathematics for decisions under uncertainty.

My education joined rigorous mathematics to computation and real systems. That combination still shapes both my professional work and my teaching.

2009–2012 · Télécom Paris

Mathematical engineering

I entered Télécom Paris after the French mathematics, physics, and computer science preparatory track. My concentration included probability, statistics, stochastic calculus, real-time embedded systems, and Malliavin calculus.

Télécom Paris is a selective French grande école whose engineering curriculum combines mathematics, computing, physical systems, and project work. The lasting lesson for me was that theory becomes more useful without becoming less rigorous when you have to implement it.

Télécom Paris engineering program

2011–2012 · Paris VI and École Polytechnique

Probability and finance

I then completed the joint Probability and Finance master's at Université Pierre et Marie Curie (now part of Sorbonne Université) and École Polytechnique, commonly known as the El Karoui program.

The program brought together stochastic calculus and control, Monte Carlo methods, statistics, numerical methods, market models, and risk. My thesis modeled participation in dark pools through a price-impact dynamic. It taught me to move between proof, simulation, data, and a decision made under uncertainty.

Sorbonne Probability and Finance program

Both programs are selective and mathematically intensive. The habits they reinforced matter most to my story: state assumptions, build the model, test it numerically, and know what the result cannot tell you.

Career

Building decision systems for electronic markets.

I have worked across FX, rates, and futures at Barclays, AIM2, and Citadel. The common thread has been systematic trading and execution: turning market behavior into models, testing those models, and building systems that can make disciplined decisions in production.

2013–2020

Barclays

Pricing, market making, internalization, execution algorithms, and relative-value systems across currencies, rates, and futures.

2020–2021

AIM2

Research leadership on reinforcement-learning systems for market making, execution, hedging, and balance-sheet decisions.

2021–2026

Citadel

Cross-asset execution research and internalization, from signal and simulation design through production implementation.

Selected work, in plain language

Three problems I found especially interesting.

The descriptions stay deliberately broad. They explain the mathematical problem without publishing trading logic, data, performance, or other proprietary details.

Learning how to choose and tune execution algorithms

I worked on reinforcement-learning approaches to an “algo wheel”: a system that chooses among execution methods and adjusts their parameters as conditions change. The challenge is to learn from results without confusing luck, market movement, and genuine execution quality.

A market maker trained with policy gradients

At AIM2, I helped develop a deep-reinforcement-learning market maker. A policy-gradient model had to balance several competing outcomes, including the chance of trading, inventory and hedging risk, and the return on the balance sheet it consumed.

Portfolio decisions across several time horizons

I applied optimal-control ideas to portfolio construction and to systems in which fast and slow decisions interact. Improving one horizon in isolation can make the whole system worse; the design has to account for those interactions.

Teaching at Primoris Academy

If I cannot explain it to a student, I probably do not understand it yet.

I teach mathematics, financial literacy, and data science to students in grades 5 through 12 at Primoris Academy , where students are placed by readiness rather than age and learning is built around doing.

Teaching keeps me honest: if an explanation cannot be inspected, changed, or challenged, it is not finished. Sometimes the right response to a blank stare is not a longer explanation. It is a different one.

Explore the courses

Outside work

Family, jiu-jitsu, music, and a business close to home.

Family

Being a father is central to why this site exists. Numeria began as a way to make summer mathematics engaging for my daughter, and many of my teaching questions begin at home: what helps a child become confident, independent, and willing to stay with a hard problem?

Brazilian jiu-jitsu

I am a blue belt and train gi and no-gi. My daughter Clémentine recently earned double gold at NAGA and silver at ADCC. I am proud of the medals, but even more proud of the habit behind them: she keeps choosing hard things and shows her younger sister what it looks like to keep going. Brazilian jiu-jitsu teaches our family decision-making under stress, honest feedback, and humility.

Our family business

Laliberte Interiors

My wife, Lisa Laliberte-Alle, is the founder and principal designer of Laliberte Interiors, a full-service residential design studio serving Northern New Jersey, Manhattan, and Connecticut. The design is hers; I help behind the scenes with software, automation, and operational systems. It has grown into a thriving family business, and I am proud to help Lisa build the machinery that supports the creative work.

See Lisa's work

Music

I have played guitar, bass, and drums since my teens and recorded three albums. Music taught me another version of deliberate practice: isolate a difficult passage, slow it down, listen closely, and put it back into the whole.

Why Dad Maths

Dad Maths is my attempt to bottle the good kitchen-table explanation: personal enough for one child, clear enough to help the next. I write down what works, what flops, and what I am still building. I also want children to reason with AI instead of letting AI reason for them.

Find me on LinkedIn