Over 5 years of experience building and deploying ML models in production. Specializing in EdTech, adaptive learning, recommendation systems, MLOps, with recent project experience in LLMs and NLP. Seeking to apply my expertise in model optimization and robust software engineering to solve challenging business problems.…

Steve Majou

PRO

Over 5 years of experience building and deploying ML models in production. Specializing in EdTech, adaptive learning, recommendation systems, MLOps, with recent project experience in LLMs and NLP. Seeking to apply my expertise in model optimization and robust software engineering to solve challenging business problems.…

Available to hire

Over 5 years of experience building and deploying ML models in production.

Specializing in EdTech, adaptive learning, recommendation systems, MLOps, with recent project experience in LLMs and NLP.

Seeking to apply my expertise in model optimization and robust software engineering to solve challenging business problems.

See more

Language

French
Fluent
English
Fluent
Spanish; Castilian
Beginner

Work Experience

Data Scientist - Adaptive Learning at Lalilo
October 1, 2018 - December 20, 2023
Designed and implemented a data pipeline for automated model training, reducing deployment time from 5 days to 3 hours. Led production launch and ongoing improvement of two new recommendation algorithms (CatBoost & ELO), improving AUC by 7%. Refactored back-end code: wrote clean, robust Python, and conducted regular reviews with Software Engineers. Collaborated on the redesign of transformer-based speech recognition system: model training, evaluation, and production deployment. Managed the adaptive-learning system redesign and performed literature reviews. Implemented 10,000+ tagged content lines to improve analysis of student answers and support data-driven EdTech decisions.
Data Scientist Intern at Lalilo
October 1, 2018 - October 1, 2018
Extracted, cleaned, and analyzed Lalilo student data; improved the existing adaptive-learning model (IRT) and implemented a new 3-parameter HIRT model, increasing AUC by 4%. Developed the new model in Python with gradient-descent optimization using TDD. Responsible for data components of student and class reports for parents and teachers; collaborated with pedagogy specialists.

Education

Master's Degree - Internet, Services and IoT at ENSIMAG (NATIONAL SCHOOL OF COMPUTER SCIENCE AND APPLIED MATHEMATICS OF GRENOBLE)
September 1, 2015 - July 1, 2018
Software engineering, algorithms (Python, Java, C, Ada) Relational databases (design and implementation) Operating Systems, Networks and network security Software engineering project: creation of a compiler for a subset of Java
PREPARATORY CLASSES at CHAMPOLLION
September 1, 2013 - July 1, 2015
Advanced mathematics, statistics, physics & chemistry

Qualifications

Kaggle Competition - Gold Medal
January 11, 2030 - November 27, 2025
Advanced Python & Classification, Regression and Neural Network Algorithms
January 11, 2030 - November 27, 2025
DataQuest, ISLR, Advanced Python & Classification Courses
January 1, 2018 - January 1, 2018

Industry Experience

Education, Software & Internet, Computers & Electronics
    paper Modeling Student Knowledge over time - Riiid! Data Science Competition

    Used an extremely simple Elo Rating algorithm to model student knowledge on an e-learning platform, which achieved higher scores than the baseline of state-of-the-art self-attentive knowledge tracing or gradient boosting algorithms (SAKT, SAINT+, LightGBM), while having significantly higher interpretability, and being much more adapted to a potential real-time production deployment.