About Raphaël
Data Scientist & Machine Learning Engineer with 8 years of experience.
I specialize in implementing large-scale ML projects from initial exploration to production deployment. Currently a Tech Lead Data Scientist at Air France-KLM, I’ve led complex data science projects, including migration to Google Cloud Platform and industrializing predictive solutions.
I’m available for freelance missions starting mid-November 2024.
With expertise in machine learning engineering and a tool-agnostic approach, I can help transform your proof-of-concept or MVP into an industrial, clean, and monitored product.
My Stack
- Programming & IDEs: Python, (py)Spark & SparkML, C++, bash, Pycharm, VS Code, Github Copilot ❤️
- Analytics & Machine Learning: JupyterLab, Vertex AI Workbench, scikit-learn, pandas, tensorflow, tensorflow-probability, xgboost/lgbm, matplotlib, plotly, streamlit
- LLMs: Langchain, OpenAI API
- MLOps & inference: Airflow, Kubeflow, MLflow, FastAPI, Docker, Kubernetes
- CI/CD: github, github actions, Bamboo, Terraform
Cloud Computing
- Google Cloud Platform: Vertex AI, Vertex AI Pipelines, Cloud Run, Cloud Functions, BigQuery
Misc.
- Training Modules: Development of training modules for Data Scientists, covering probabilistic ML, Python best practices, MLOps, and more
- Mentoring & Coaching: I offer mentoring and technical coaching sessions for juniors or career changers
- Best Practices: I can audit your code, propose best practices, and provide templates to improve your workflow.
I’m a great fit for you if:
You promote inclusivity 🏳️🌈
You have a culture of sharing
You are innovative and strive for technical excellence
Let’s Collaborate!
If you’re ready to take your data science projects to the next level, I’m here to help.
French
Native or bilingual
Can work on-site
Paris (up to 15km)
Experience
- Air FranceTech Lead Data ScientistTRANSPORTATIONNovember 2022 - Today (3 years and 7 months)Bd Périphérique, Paris, FranceTech Lead Data Scientist – Ground Operations Team (Part Time)• Tech Lead Data Scientist for a set of transversal products related to on-time performance, flight schedule robustness, and flight delay predictions• Updated and re-architected a 2019 PoC to meet coding standards, led its industrialization and model deployment, and improved runtime performance• Migrated, as a front-runner, this ML application to Google Cloud Platform, including the MLOps training pipeline (Vertex AI Pipeline) and the infrastructure using Terraform, in collaboration with Ops teams• Supervised 3 junior data scientists on technical aspects and mentored an internship project on ML probabilistic techniques applied to the project• Techniques used: Monte Carlo simulations, quantile regression (LightGBM), probabilistic neural networks, Vertex AI Pipelines, TerraformTech Lead Data Scientist – ORTech (Part Time)• ORTech is a multidisciplinary team supporting ML project industrialization, best practices, and workflow standardization within the department• Participated in the design and implementation of a standard MLOps stack for the department• Developed and facilitated technical training modules for data scientists, including pytest, probabilistic ML, clean code, and Vertex AI Pipelines• Supported product teams on architectural, implementation, and GCP cloud migration concerns
- MP DATASenior Data ScientistDIGITAL AND ITFebruary 2021 - November 2022 (1 year and 9 months)Paris, FranceTech Lead Data Science Ground Operations @AirFrance- Data Exploration, feature engineering, model training & hyperparameter tuning- Architecturing machine learning projects (Short-term passenger prediction, flight delay) for on-prem and cloud-ready environments- Building end-to-end pipelines (training+inference) in docker/k8s environmentsSenior Data Scientist - Operation Research Tech Team @AirFrance- Proof of concept/value on MLOps solutions, including Kubeflow, MLflow, data-drift monitoring (Kafka/ELK/Kibana stack)- Hyperparameter tuning tests with kubeflow-katib- Building kubeflow pipelines with kubeflow-kaleTechnical Comitee Member @MPData- Data architecture reviews, technical job interviews- Call for tenders: case studies for a parisian public transport operaator, machine learning kaggle-like study (Time series prediction, involving prediction on covid period)
- MP DATAData ScientistAugust 2017 - October 2021 (4 years and 2 months)Paris, FranceAugust 2018 - March 2020 Predictive Maintenance - Data Science Consultant - Paris Region I was involved, together with a 15-member team, in an aircraft predictive maintenance project at a big French airline.- Developping ML Models (from the initial research phase to the productisation), mostly CARTs and unsupervised outlier detection methods (Local Outlier Factor, One-Class SVM)- Presenting the results to business, with a focus on adapting the speech for the analysts' mindset- Adapting and deploying the "data treatment application" to the cloud, for the clients' partnerships with other airlines.- On a more ML Engineering aspect, I deployed the application on an on premise Hadoop infrastructure for one of the client's subsidiary Cloud stack: EMR, AWS Lambda, S3, DynamoDB, Sagemaker Notebooks, boto3, aws-cli General stack: Python, Pycharm, Jupyter Notebooks, sklearn, seaborn/ matplotlib, MongoDB August 2017 - August 2018 Crew Scheduling - Operation Research Engineer Consultant - Paris Region For a big French airline, I participated in the "Crew Scheduling" project.- Implementing LP engines for activity affectation to cabin crew officers- Maintaining/developing, a so-called "rules engine", to ensure a planning fits the regulation (e.g. the law, union agreements, ...) Stack: C++, CPLEX
Recommendations
Be the first to recommend Raphaël
Help this freelancer shine by sharing your experience working together.
These freelancer profiles also match your criteria
Agatha Frydrych
Backend Java Software Engineer
4.7
(3)
2
Baptiste Duhen
Fullstack developer
4.6
(4)
5
Amed Hamou
Senior Lead Developer
4
(2)
7
Audrey Champion
Web developer
4.3
(3)
4
Education
- Management & Entrepreneurship Certificate, Entrepreneuriat / études entrepreneurialesISAE-SUPAERO2016Management & Entrepreneurship Certificate, Entrepreneuriat / études entrepreneuriales
- Diplôme d'ingénieur, Decision Sciences - Financial EngineeringISAE-SUPAERO2016Stochastic processes, time series, derivative pricing, market finance, corporate finance