À propos de Mehdi
Français
Bilingue ou natif
Anglais
Capacité professionnelle complète
Arabe
Bilingue ou natif
Expériences
- NIELSENIQDATA ENGINEERfévrier 2022 - Aujourd'hui (4 ans et 4 mois)Paris, FranceProject 1 - Semantic Search Engine :Build a search engine based on vector data basePrepare and clean Data.Upload data into the vector data base (Qdrant).Build API interface (FastAPI).Project 2 - Image Processing (ETL / ML Project):An end to end solution, it allows the clients to compare theauthenticity of images of their products published by their retailers.Database modeling (Cloud SQL).Build ETL pipeline (GCP Workflows, Cloud Run, Cloud SQL).Setup CI/CD pipeline (Bitbucket, Cloud Run, Cloud Build, Secrets).Unit Tests (Pytest, Tox)Build and Deploy ML modele (Kubernetes).Manage junior Data Engineers on the project.Project 3 - Promo ExtractionMachine learning (R&D) project that aim to extract relevant informations from promo textPrepare and clean Data.Compare ChatGPT, Gemini to propose relevant labels for NER models.Testing the recent model GLiNER.Project 4 - Product ProcessingBuild an ETL to process product data from scrapping spidersBuild ETL pipeline: MariaDB, Cloud Run, Polars, MongoDB, Airflow.Build APIs to provide data for different teams to get product data (Streamlit, Mongo).Setup CI/CD pipeline on Bitbucket.Technologies : Python, Django, Polars, MongoDB, Cloud SQL, SQL, CloudRun, CloudFunction, Docker,Kubernetes, Bitbucket, Tox, Streamlit, NoteBook, ChatGPT, Gemini, Qdrant, FastAPI
- DcubeData Engineerjuin 2020 - février 2022 (1 an et 8 mois)Paris, FranceProject 1 - Prediction of the number of sales per storeMission (POC) to present the Dataiku solution through a business use case for predicting the number of sales per storeLead the brainstorming session with the client.Build dashboards on Dataiku.Training a Linear Regression model.Project 2 - Data warehouse migration project to the Azure Cloud :Integration of new data flows and building ETLs pipelines on the Azure cloud.Define the pipeline and the different final users of the data.Build data flow on Azure.Upload data from on-premises servers to the Azure DataLake.Create CI/CD pipelines on Azure DevOps.Technologies : Python, Azure Datalake Gen2, Azure DevOps, Windows, Dataiku, Azure, Python, SQL
- Constant oatsINVENTIV ITjanvier 2019 - mai 2020 (1 an et 4 mois)Project 1 - Job Matching Platform :A solution built from scratch Allows recruiters to matchs the received CVs with the job offer descriptionImplement a CV recommendation system, Data collection (open API, CVs, HR database,Scrapping)Design of an ETL in batch mode.Implement and deploy the TF-IDF modelProject 2 - Named Entities ExtractionEnsure a better user experience through the development of a Deep Learning model for the extraction of named entitiesCollect new datasets.Build Machine learning models.Technologies : Python, Pandas, Sk-learn, Postgres, Flask, Lambda Function, S3, AWS, Python, Azure Datalake Gen2, Azure DevOps
Recommandations
Soyez le premier à recommander Mehdi
Contribuez à la réussite de ce freelance en partageant votre expérience de collaboration avec lui.
Ces profils de freelance correspondent également à vos critères
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
Formations
- Master of International StudiesUniversite esn Monnet de S-Etenne,2019MASTER INTERNATIONAL MACHINE LEARNING ET DATA MINING
- DIPLOME D'INGENIEUR INFORMATIQUE LASSES PREPARATOIREeda2013DIPLOME D'INGENIEUR INFORMATIQUE LASSES PREPARATOIRE