Freelancer profile translated to English.
Back to original languageAbout Raphaël
I primarily develop in Java (7 years) recently for exposure APIs, and more particularly in Scala (4 years) for Spark and console applications. One must be a JVM polyglot these days.
Initially using V-methodology, I now work with a mix of Scrum + Kanban for transparency and the Agile Manifesto as a philosophy.
To make an application sustainable, I promote collaborative design.
I am appreciated for my commitment to projects, my assertiveness in exchanges, and my pragmatism in design and development.
After holding positions as Consultant, Integrator, Java Developer, Big Data Developer, and Big Data Lead Developer in various sectors such as Mass Retail, Energy, Finance, and Transport, I wish to continue assisting companies as a facilitator of their Master Data Management transformation.
It is to find assignments exclusively within this framework that I created the company Flulo, the principle is the same as that followed by Google, Amazon or Facebook, to use Open Source frameworks to have the most reliable and Low Cost Master Data Management possible.
The evolution towards horizontally scalable architectures and specific developments based on open-source frameworks should be the target of any company as it is the best way to reduce and control hardware and maintenance costs.
Furthermore, if we combine this with a de-siloting of organizations and a holistic approach, we find extremely beneficial shortcuts to productivity. (such as real-time report creation).
English
Fluent
French
Native or bilingual
Can work on-site
Noisiel (up to 10km), Noisy-le-Grand (up to 10km), Chessy (up to 10km), Paris (up to 10km)
Experience
- EnergismeHbase/Phoenix Expert Trainer and CoachENERGY AND UTILITIESApril 2021 - Today (5 years and 2 months)Boulogne-Billancourt, FranceTransition from a state of blockage to a series of quick wins. The team now shares a simplified vision of the target architecture and gains mastery with each new step.Details:• Facilitation of an introductory day on Hbase/Phoenix• Consideration of performance issues and multiplication of treatments• Workshop to solve a hot-spotting problem using few cluster regions (+3000% read performance and -70% disk resource usage)• Participation in brainstorming meetings for roadmap refinement• Proposal of architectural simplifications and regrouping/reduction of treatments• Assistance in designing a more efficient Bulk Loading process using Spark.• Demonstration of the performance of the map reduce job allowing the ingestion of 1 million lines in 30 seconds.
- FNAC DartyTechno-Functional ConsultantRETAIL (LARGE RETAILERS)October 2014 - February 2016 (1 year and 5 months)Paris, France• Outcome: Establishment of a complete organization from scratch for the creation of a unique Product Repository for Fnac's Marketplace, enabling it to withstand competition from Amazon.• Keys to Success: The industrialization of deployments with Gradle transformed our business and led to the project's success. We went from 90% integration and 10% development to 90% development and 10% integration.Details:• Response to Request for Proposal• Proof of Technology feasibility demonstration• Refined Proof of Concept feasibility demonstration• Leading technical interviews• Training on MDM CE11 / Java EE / Integration• Participation in requirements gathering workshops• Writing the General Functional Specification• Writing Technical Specifications• Creation & Maintenance of the data model• Design integrated with MDM CE 11• Deployment industrialization (less than 10 minutes, with Gradle + ANT + Shell Script)• Automation of external DB2 SQL table deployments in non-ISO environments• Development of specific UIs: bidirectional links, asset metadata, initial variant model (with AJAX)• Estimation of the workload for Business Rules development• Task allocation• Progress tracking with a shared tracking file and daily team stand-ups• Writing the deployment procedure
- EDF CommerceTech Lead Big DataENERGY AND UTILITIESApril 2018 - June 2019 (1 year and 3 months)Paris, FranceInitial mission of feeding and exposing reference data:• Self-organization around a Kanban with Daily Scrum• Planning poker for breakdown into 2-week User Stories• Hortonworks Platform: HDFS, Kafka, Spark, Hive, Hbase, Ambari, Yarn• Adaptation: Initialization with Wip Limit at 1 to chain Stories through team focus• Adaptation: Automations with Gradle and Jenkins• Adaptation: Creation of a complete development environment installer for the entire division• Adaptation: Demonstration with an Angular JS project to share with other teams• Evolution: Support and accelerator role for production teams• Adaptation: Test First with Cucumber Framework and Behavior Driven Development• Adaptation: Learning Scala for Hbase ingestion APIs with Spark Streaming• Adaptation: Creation of Hbase data exposure Web Services with Jax-RS (Jersey2)• Adaptation: Creation of a performance and load testing project with Gatling• Adaptation: Exposure API made generic• Outcome: Creation of the repository for energy consumption and billing, and exposure APIs necessary for marketing supervision and EDF applications on your smartphones.• Keys to success: The organization of small iterations around a board representing the steps necessary for rapid production deployment allows for a shared vision. (Design>Test>Development>Deployment)The code for stream and batch processing, as well as for the exposure APIs, was as generic as possible, which allowed us to shift our activity from 90% development 10% integration to 40% Big Data architecture upskilling, 50% development, 10% integration.Details:• Adaptation: Industrialization of development processes in each project: Cucumber tests with reports, hot debugging, multi-environment delivery bundle...• Training: Dissemination of best practices through pair programming within the team and through technical workshops with the rest of the division• Adaptation: Takeover and improvement of the generic API exposure creation project (Gradle+Spring Boot) for Hbase repository data (code generation in contract first with Swagger 2.0)• Adaptation: Creation of a pure Spark+Scala ETL codebase (Hbase -> Elastic Search) dedicated to cross-search indexes and API pagination. It creates indexes in 3 to 10 minutes instead of 1 to 3 hours with the old solution.• Enrichment and maintenance of a recurring project with Spark Streaming (Kafka -> Elastic Search)
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
- Java EE Application Designer Developer (RNCP level 6)AFPA2014Projet : développement d’un site d’e-Commerce Librairie en ligne Back Office avec Swing + JBDC + SQL Server 2008 Front Office sur modèle M.V.C avec pour Vue JSTL + CSS et un Contrôleur Servlet Durée : 1 mois
- Higher National Diploma in Management Information Technology, Application Developer optionMicropole Institute Paris2005Projet : développement d’un site de Gestion de la Relation Client GRC pour une société de Télémarketing (Front Office) Durée 1 mois
Certifications
- IBM InfoSphere Information Server for Data Quality Fundamentals Technical Professional v1IBM Training2014