07/2022 - Current

Internal Missions

• Application of state-of-the-art algorithms and techniques in Reinforcement Learning for electric power distribution systems.
• Research and innovation related to the application of Graph Neural Networks to electric power distribution modelization.
• Contribution to the development of web applications, solutions, and APIs.
• Development of AI generative tools, such as chatbots, RAG, summarization, and others.
• Participation in responding to tenders. & creation of PoC and Demo applications.
• Supporting junior data scientists in their missions & Contribution to the development of internal data expertise.

External Missions

• AMOA & DevOps role in the development of a web application related to budget management.
• Patronage in IT and data-related subjects.

Tools Python, PyTorch, PyTorch-geometric, Falsk, Dash, RestAPI,Ninj-API, LLM, RAG, Ollama, Langchain, llamaindex, VectorDB, Fine-tuning, quantization, OpenAI, Streamlit, Docker, CI/CD, MLOps

Skills Deep Learning, Reinforcement Learning, Graph Neural Network, NLP, Backend Dev, API

01/2021 - 07/2022

• Managing and exploiting data from various sources, including structured and unstructured data, to generate valuable insights.
• Creating and producing reports and indicators that provide clear and concise information to stakeholders for decision-making purposes.
• Designing and developing machine learning (ML) and deep learning (DL) models for various business challenges and provide innovative solutions to complex problems.

Tools Python (Numpy, pandas, tensorflow, keras....), R, Shiny, Agile, Jenkins, GitLab, Jira, Docker

09/2020 - 01/2021

• Project plan, database preparation, data processing and structuring
• Choice, development and improvement of ML models.
• Industrialization and monitoring of ML models.

Tools R, Rstudio, Python Anaconda, SQL, SAS / Informatica, JSON, Dataiku, qlikview, AzureML

03/2020 - 09/2020

Development of an experimental tool based on the use of Machine Learning in medical imaging in order to provide a diagnostic aid tool.

• Performed lesion segmentation using U-net model and radiomics feature extraction on MRI data.
• Developed and compared multiple machine learning models for lipomatous soft tissue tumor classification, including LR,SVM,RF,GB, and DL models.
• Evaluated and compared model performance using k-fold cross-validation.
• DeLong's test, and various performance metrics, ultimately obtaining the best results with a batch-corrected radiomic data GB model.

More about this paper can be found here

Tools Python, Jupyter, Tensorflow, Scikit-Learn, Numpy, Keras, Pandas, Scipy, tkinder, Nvidia DGX Linux, Docker, Shell ,slurm

10/2018 - 03/2020

Creation of contextual service applications within the vehicle.

• Development of applications of data collection from the vehicle.
• Data collection and pre-processing (filtering, data wrangling…),
• Choice, development, test and validation of ML Models
• Implementation and validation of POCs carried out

Tools Python (Scikit-Learn, Numpy, Keras, Theano, TensorFlow, NLTK, Pandas, Scipy, t kinder, Cantool, multiprocessing), Apache Spark, R, Jupyter, AWS (EC2, S3, CLI, EMR), Linux, NVIDA, CAN bus, JSON, CSV, geopandas, OpenStreetMap Overpass API (osmnx)

04/2018 - 10/2018

• Big Data Setup: Tool configuration, data extraction, and preprocessing.
• Feature Selection: Identified study populations and extracted key characteristics.
• Algorithm Evaluation: Compared unsupervised classification methods.

Tools Python, Pyspark,Hbase, HDFS, PowerBI, Jupyter, Linux

Skills Deep Learning, Reinforcement Learning, Graph Neural Network, Backend Dev, API

01/2014 - 10/2017

Leading industrial process implementation, automation and optimization.

• Planning and executing process development and implementation.
• Coordinate and manage activities, resources, budget and schedule.
• Optimization and improvement using AI and algorithmic:
• Prediction of the best optimal parameters for the welding machine by NNs (Theano)
• Find the most suitable raw material alternative for terminals in the event of a shortage of RM (Clustering, Python)
• Automated quality control and anomaly detection: (AlexNet, Keras)
• Prediction of order forecasts for RM supply and budget, space and equipment allocation (ARIMA)

Tools: Python, Theano, Keras, Linux, GSD, SAP, Excel, VB

Skills: Six Sigma, Process Improvement, Anomaly Detection,Deep Learning, Product Management