Freelancer profile translated to English.
Back to original languageAbout Raphael
👋 Welcome to my Malt profile! 🚀
I'm Raphael,Data Engineer,**DevOps**, **CI**, and **Software Engineering**, specializing in **Python**. With over 7 years of experience, I help companies optimize their data and software development processes with tailor-made solutions.
Here are the missions I regularly undertake:
🧑💻Data Engineer&DevOps🧑💻
- **Data Engineer**: Performance optimization
- **Data Engineer**: Corrective and evolutionary maintenance
- **Data Engineer**: Debugging
Software Engineering 💻
Some of my achievements:
🏦 **Allianz Trade**: Deployment of a data preprocessing system in **Python**, increasing the accuracy of risk predictions.
🤳Bloom**: Implementation of a preprocessing platform with **Kafka**, handling over a million documents for faster insights. Micro-services in **PythonandGo
My current stack:
📈 I aim for excellence, ensuring that every solution is robust, scalable, and meets the needs of each client.
Looking for an expert in **Data Engineer**, **DevOps**, **CI**, or **Software Engineering**?
I respond **in under an hour**!
French
Native or bilingual
English
Fluent
Can work on-site
Paris (up to 50km), Lyon (up to 10km)
Experience
- Allianz TradeDate EngineerBANKING AND INSURANCESeptember 2022 - Today (3 years and 11 months)Paris, FranceAs part of a strategic project within theCredit Modeldepartment, the team had to develop anexplainable algorithm(logistic regression or EBM type) to ensuretransparency and compliance.I contributed to this mission by participating indata cleaningand setting up arobust ETL pipelineto feed this model.
- Integration into adynamic teamof Data Scientists as a **Data Engineer**, contributing to data cleaning and preparation for model training.
- Design of apreprocessing servicein **Python**, deployed with **FastAPI**, dedicated to data optimization.
- Complete implementation of a **CI/CD approach**, including **tests**, code formatting, **security checks**, and deployment stages.
- Restructuring of existing code to optimize it and adapt it to current development standards.
This model wassuccessfully deployedin several countries within the group, and the results obtained were very positive, confirming the value of the implemented approach. - BloomData EngineerSOFTWARE PUBLISHINGApril 2021 - September 2022 (1 year and 5 months)Levallois-Perret, FranceAt Bloom, a company specializing insocial network analysisfor security, defense, and marketing, I worked onimplementing an ETLbetween data scraped from different platforms (Facebook, Twitter, etc.), processing it in **Python/Kafka**, and then loading it into a graph-oriented database for later use by the Data Scientists team.
- Implementation of astreaming platform**, covering agile development, creation of a multilingual tokenizer, and integration of **Prometheus metrics.
- Reliable storage of data fromKafka to S3**, with deployment on **Kubernetes automated by GitLab CI.
- In-depth expertise inGit/GitLabtools: managing clean commits, rebases, rigorous code reviews, and continuous integration.
- Team collaboration: pair programming, daily meetings, **structured backlog management**, and technical workshops.
- Clean coding: strict application ofTDD**, the **KISSprinciple, and regular refactoring to maintain quality.
- Structuring of a V2 based onKafka micro-servicesinPython via gRPC**, **Docker containerization**, and centralized monitoring with **Prometheus/Grafana.
- **Performance optimization**, scalability management, and resource consumption control.
- Refactoring of theclassification algorithmand automation of data loading toAWS S3.
- Approach combining technical rigor, collaboration, and innovation.
This V2 of the platform demonstrated its effectiveness and robustness, but was eventually replaced by a more performant **Spark-based ETL**, better suited to the operational needs of the tool. - SouthpigalleData EngineerSOFTWARE PUBLISHINGSeptember 2018 - April 2021 (2 years and 7 months)Paris, FranceAtSouthpigalle(now part of the **Meytis group**), I worked on the design and deployment of architectures dedicated to **NLU**.This startup specialized inNLPexploitation to enable companies to access and leverage customer data.
- Design of anNLP micro-services architecture**, covering the entire chain: from **preprocessingtomachine learning algorithms.
- Close collaboration with Data Scientists for theimplementation of a sentiment analysis modelbased on **BERT**, and integration of advanced language preprocessing libraries.
- Development of specializedmicro-servicesfor NLU, communicating via aGunicorn/Flask API**, allowing dynamic enrichment of **JSON objects.
- Creation of anExpert-System Genius**, exploiting **NLPto automatically convert natural language sentences intoGraphQLqueries.
- Collection and integration of user feedback, deployment, and maintenance of all services on Kubernetes.
- Approach based on **technical rigor**, co-construction with the team, and constant innovation.
I participated in the implementation of a high-performance NLP model, which was successfully used in several concrete projects for prestigious clients in theluxuryandfood industries.
Reviews
Recommendations
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
- Digital Technology Architect42201742, c’est une formation en informatique d’excellence pour tous et toutes. Au programme de cette école différente et innovante : une approche par projets pour progresser et développer des compétences techniques et humaines recherchées sur le marché du travail.