About Me

My Story

I’m Nassim Louissi, a Machine Learning Engineer based in the vibrant city of Paris. My journey into ML began during my Bachelor’s in Economics at University Paris-Est Créteil, where I discovered the power of data and statistics. This passion led me to pursue a Master’s in Econometrics and Statistics at the University of Angers, focusing on machine learning applications in healthcare and beyond.

At Quinze-Vingts Hospital, I’ve developed AI models that make a real impact, from predicting eye diseases to optimizing pipelines. Outside of work, I enjoy brewing coffee, exploring Paris’s hidden trails, and staying updated on the latest in AI. I’m always open to collaborations on innovative projects—feel free to reach out!

Experience

  • Machine Learning Engineer (Initially joined as Intern)
    Quinze-Vingts Hospital, Paris, France
    April 2024 - Present
    • Built AI models (ResNet, CNN, ViT, CatBoost) in Python/PyTorch to predict eye diseases, achieving 97% accuracy on 1,000 samples and 100% success on 120 samples (vs. 75% baseline), saving €70k-120k/year.
    • Reverse-engineered imaging hardware (MS39/Corvis-ST, 150k exams) using optimization algorithms, achieving 2-6% deviation for computer vision.
    • Optimized ETL pipelines to extract/calculate 4,000 features at 5x speed (30 samples/min) for Gradient Boosting inputs.
    • Deploying models using Docker/Kubernetes; collaborated with clinicians for production and validated features/metadata integration.
    • Published 1 paper (IF~2) as co-first author and drafted 2 more on ML/statistical modeling for surgery outcomes.

Education

  • Master of Science in Econometrics & Statistics
    University of Angers
    September 2023 - September 2025
  • Bachelor in Economics
    University Paris-Est Créteil
    September 2019 - September 2023

Projects

  • Hackathon ElevenLabs x a16z
    LLM Medical Assistant
    Developed an LLM-based Medical AI Assistant for a Hackathon (ElevenLabs x a16z) to guide medical diagnosis with text and voice capabilities using ElevenLabs API, tailored for populations without healthcare access; built full-stack architecture with Next.js, React.js, and Tailwind, deployed on Vercel for production.
  • Real-Time E-commerce ML Pipeline: Product recommendation system
    Built a scalable real-time product recommendation system using collaborative filtering.
    Deployed on AWS with Airflow and Kubernetes, automating CI/CD for model training/evaluation/deployment and integrating CloudWatch/Prometheus for continuous monitoring and performance.

Skills

  • Programming Languages: Python, SQL, Javascript, Java
  • Cloud Platforms: AWS (S3, SageMaker, Lambda, EC2)
  • MLOps Tools: Docker, Kubernetes, Airflow, Terraform, MLFlow, Prometheus, CloudWatch
  • Machine Learning Frameworks: PyTorch, TensorFlow, Scikit-learn
  • Machine Learning Libraries: Pandas, Polars, NumPy, SciPy, OpenCV
  • ML Techniques: Computer Vision, Time-Series, Conformal Prediction, NLP (Transformers, LoRA)
  • Languages: French (Native), English (Bilingual – C1)

Publications

  • Perez E*, Louissi N*, et al. "Machine Learning Model for Predicting Visual Acuity Improvement After Intrastromal Corneal Ring Surgery in Patients With Keratoconus." Cornea, 2025. (Co-first author)

References

Head of Service & Chair of the Medical Establishment Committee, Quinze-Vingts Hospital

Available upon request