Experience

  1. PhD Researcher

    Institut Curie & Sorbonne Université
    • Developed MIIC_search&score, a novel search-and-score algorithm for causal graph discovery from nonlinear and non-Gaussian data with latent variables, implemented in R.
    • Designed and automated a large-scale benchmarking pipeline comparing five causal discovery methods across 960+ simulated graph configurations, executing thousands of parallel experiments on HPC clusters (SLURM, PBS) with datasets containing up to 20,000 samples.
    • Improved graph recovery precision by 18% on continuous data and 27% on categorical data compared with the MIIC baseline algorithm, leading to a first-author publication at ICML 2025.
    • Contributed to additional publications in computational biology (eLife) and digital health (npj Digital Medicine).
    • Contributed to the development and maintenance of the open-source MIIC R package and developed MIIC-Display, an interactive web application for causal graph visualization using JavaScript, D3.js, PHP, and SQL.
    • Mentored two master’s research interns on research projects in causal discovery.

  2. Data Analysis Consultant

    Sorbonne Université
    • Developed Python workflows for cleaning, analyzing, and visualizing 100+ quantitative and qualitative survey responses.
    • Designed reproducible statistical analyses and visualizations to evaluate the impact of a university pedagogy initiative.
    • Produced reports that informed decision-making and contributed to a publication at QPES 2025.

  3. Teaching Assistant

    Sorbonne Université
    • Delivered 80+ hours of practical instruction in Python, C programming, and biological network inference to undergraduate and master’s students.
    • Supervised master’s student projects.

  4. Research Intern (M.Sc. Thesis)

    I2BC & Université Paris-Saclay
    • Developed a reference-free RNA-seq pipeline, ksub, for tumor-specific transcript detection from large-scale sequencing data, implemented in Python and C++.
    • Processed 1.4 TB of RNA-seq data (17 billion reads, 213 samples) and constructed large-scale k-mer indices containing up to 494 million k-mers.
    • Designed a multi-stage filtering strategy combining graph-based assembly, k-mer indexing, and negative binomial modeling, reducing the candidate transcript space by 99.85%.
    • Integrated an end-to-end bioinformatics workflow including Cutadapt, FastQC, BCALM2, BLight, REINDEER, STAR, and BLAST for preprocessing, indexing, assembly, alignment, and annotation.

Education

  1. PhD in Computer Science

    Sorbonne Université
    PhD thesis: Learning Ancestral Graphs via a Search-and-Score Approach
    Research conducted at Institut Curie in Paris.
    Supervisor: Dr. Hervé Isambert.
    Funded through the Imperial – CNRS Joint PhD Programme on Digital Transformations and Global Challenges.
    Read Thesis

  2. M.Sc. in Bioinformatics & Modelling

    Sorbonne Université
    Master thesis: Towards a Complete Molecular Portrait of an Individual Tumor Using RNA Alone
    Research conducted at I2BC in Gif-sur-Yvette.
    Supervisor: Prof. Daniel Gautheret.
    Specialized in machine learning, bioinformatics, biological networks, computational neuroscience, biomathematics, and graph theory.
    Master Thesis

  3. B.Sc. in Life Sciences

    Sorbonne Université
    Broad scientific training spanning biology, mathematics, and computer science.
Skills
Programming Languages
Python
R
C++
C
Machine Learning
PyTorch
scikit-learn
Tools & Infrastructure
git
Linux
bash
HPC (SLURM, PBS)
Languages
100%
French (Native)
80%
English (Fluent)