Data scientist Amiens : Trouvez vite les meilleurs freelances disponibles

Exemple de missions de Derrick,
Data Scientist freelance

Experience

Menard Group - World Specialist in Ground Improvement Orsay, 91
Senior Data Scientist Jan. 2023 – Now
Predictive maintenance to detect and prevent anomalies on the machines
* Collecting production data: torque pressure, pump pressure and rotating speed.
* Concatenating data on each sensor to create a time series.
* Anomaly detection with the IQR: interquartile range, and the isolation forest techniques.
* Integrating acceleration data and frequence detection for the wireless vibration monitoring system.
* Time series data visualization with Seaborn
Technologies: Python 3.10 – Scikit Learn – Pandas – Numpy – Seaborn – Jupyter Lab – Windows

CETIM - Technical Center for Mechanical Industries Senlis, 60
Senior Data Scientist Sept. 2022 – Dec 2022
Analyzing the foundry parameters that impact the quality of produced pieces.
* Detecting features that impact the most the quality of produced pieces.
* Feature importance analysis with Machine Learning algorithms: Random Forest,
Logistic Regression, GradientBoosting, Permutation Feature Importance
* Boostrap Sampling to manage the unbalanced dataset
* Data Science Evangelist, Training and Vulgarization
Technologies: Python 3.7 – Scikit Learn – Pandas – Numpy – Data Visualization with Seaborn –
Jupyter Lab – Windows OS

Total Energies - Artificial Intelligence Observability ˆIle-de-France, 92
Senior Data Scientist Dec. 2021 – Aug. 2022
Comparing solutions and technologies for the implementation of the Artificial Intelligence Observability.
* Comparative study of different A.I. Obs solutions, such as Elastic Observability.
* Designed and implemented the architecture for collecting system logs from linux servers and
windows servers.
* Visualization, storage and analysis of the logs collected for the predictive maintenance of
breakdowns on the server warehouse.
Technologies: Python 3.7 – Scikit Learn – Tensorflow – NLP – Docker – Data Visualization with
Seaborn – Windows OS – Serveur Linux – Syslogs – Elasticsearch

- Consultant Data Scientist
Vallourec - Smart Tubular Solutions, Project Autograph Hauts-de-France, 59
Expert Data Scientist - ML models & Data pipeline Sept. 2019 – Auj.
D´eveloppement d’une application d’aide `a la d´ecision afin de d´etecter les mauvaises connexions entre deux tubes.
Permettant ainsi d’empˆecher les fuites d’hydrocarbures lors des op´erations sur une plate-forme p´etroli`ere.
* Conception et mise en œuvre de l’architecture de pipeline de donn´ees
* Impl´ementation d’un sch´ema de donn´ees SQL afin d’acceuillir des donn´ees de diff´erentes sources.
Un moyen de standardiser, nettoyer et contrˆoler la qualit´e des donn´ees d’entr´ee, en accord avec le
catalogue Vallourec.
* Encapsulation de ses donn´ees dans une base de donn´ees orient´ee document: MongoDB
* Calcul des features et pr´eparation des donn´ees `a l’apprentissage et `a la classification.
* Algorithmes: K-means, RandomForestClassifier, et BalancedRandomClassifier
Technologies: Python 3.7 - Scikit Learn - Imbalanced Learn - MongoDB - PostgreSQL - Rest API

Maillance - Artificial Intelligence for the Oil and Gas Industry Paris, 75
Senior Data Scientist - Artificial Intelligence Dec. 2018 – July 2019
* Analyse des intervalles de pr´ediction afin de m´esurer les incertitudes sur les features, et leur
impact sur le mod`ele pr´edictif
* M´esurer la pr´ecision du mod`ele `a l’aide la de la validation crois´ee KFold - Cross Validation.
* Calibrage des hyper-param`etres par ´echantilllonage Monte Carlo et la m´ethode GirdSearch.
* Extraction des donn´ees Log ASCII Standard (LAS): logs de donn´ees en g´eophysique et g´eologie
Technologies: NumPy - Scikit Learn - Python - Matplotlib - Lasio - Keras - TensorFlow - LSTM

Soci´et´e G´en´erale - Grid Computing Fontenay-sous-Bois, 94
Data Scientist - GPU Benchmarking Apr. 2018 – Dec. 2018
* Temps d’apprentissage du dataset CIFAR-10 sur les GPUs: K40, Tesla V100 & Tesla P100.
* Apprentissage avec Tensorflow 1.10, PyTorch 0.4.1, et Keras 2.2.0
Technologies: Tensorflow - Keras - Cuda - PyCuda - Python 3.6 - Docker - OS Red Hat 6.7 - Git

Thales Group Antony, 92
Data Scientist - Analyse des donn´ees de plan de vol Jan. 2017 – March 2018
* Pr´eparation et contrˆole des donn´ees statiques (coordonn´ees radar et plan de vol) requises par le
syst`eme de gestion du trafic a´erien (ATM) pour contrˆoler le trafic a´erospatial en temps r´eel
* Impl´ementation de la fonctionnalit´e permettant de dispatcher simultan´ement sur diff´erents centres
les donn´ees de plan de vol
* Impl´ementation de nouvelles fonctionnalit´es graphiques pour un syst`eme de transport a´erospatial:
augmentation des capacit´es du syst`eme et am´elioration de la disposition de l’interface
Technologies: Ada Language - Unit Tests - SSH connection - UNIX VM - Shell command

Experience - Academical & Internship
Applied Mathematics - Teaching & Research Draguignan, 83
Mixed Integer Non Linear Optimization Oct. 2011 – Dec. 2016
Gaussian Based Mixed Optimization to solve the Product Family Design paradigm
* S´election des composants et des variables devant constituer la plateforme de produit `a l’aide des
algorithmes de clustering tels k-means et k-nearest neighbor
* Benchmarking et comparaison des r´esultats avec ceux du solveur SCIP (Solving Constraint
Integer Programs)
Technologies: Python 3.6 - NumPy - SCIP solver - and its Python interface PySCIPOpt - MIQP

Ecole Nationale Sup´erieure des Mines de Saint-Etienne Saint-Etienne, 42
ICM Science and Engineering, Applied Mathematics Sept. 2008 – Sept. 2011
* Analyse et Simulateurs : Machine Learning - Data Mining - Probability and statistics
Data analysis - Random and Gaussian process - Martingales and Brownian motion
* Ing´enierie des syst`emes informatiques : Programmation Orient´ee Objet - UML et Design
Patterns - Gestion de Projet - High Performance Computing HPC
IFP Energies nouvelles Rueil Malmaison, 92

Data Scientist in Reservoir Engineering, Internship Dec. 2010 – Sept. 2011
Hydrocarbon recovery optimization under uncertainties
* Callibrage des hyper-param`etres du mod`ele de krigeage (m´ethode d’approximation `a l’aide de
noyaux gaussiens). Comparaison des algorithmes CMA-ES, BFGS et L-BFGS
* Determination de la position optimale o`u rajouter un nouveau puit de p´etrole: impl´ementation de
l’algorithme EGO, Efficient Global Optimization
Technologies: R - C/C++ languages - CougarFlow Reservoir Simulator
Laur´eat aux Olympiades Panafricaines de Math´ematiques Sept. 2005 – March 2006