Rachid Bechir

As part of the Digital Factory at the Accor group, I work as a full stack developer on the Accor ALL.com Mobile / Web application, specifically in the BFF (Back For Front) team. The BFF team is responsible for: - The overlay of the Accor Information System to simplify usage on the ALL.com web and app fronts. - Simplifying and standardizing API consumption by the fronts. - Aiming to be the sole entry point for the fronts regarding Accor data. Achievements: - Created an API using NodeTS with GraphQL and Apollo following the Hexagonal architecture. - Implemented a DataLoader module to optimize data fetching through the GraphQL layer, integrating batching, caching, error handling, and performance optimization with monitoring, input validation, and secure data access management. - Developed an AWS Lambda function to process request messages sent to an API exposed via AWS API Gateway for generating digital room keys, queued in AWS SQS, with notifications to users via AWS SNS, all monitored by AWS CloudWatch and secured by AWS IAM. - Pre-loaded cache on AWS Redis by creating Lambdas to fetch and cache data via SQS, with error handling through a Dead-Letter Queue (DLQ). - Implemented pre-developed Terraform modules by the infrastructure teams, including configuration, variables, data sources, planning, application, version control, Secret Manager, and IAM policies. - Utilized design patterns (Builder, Adapter, etc.) and adhered to SOLID principles. - Created a frontend application using React.js and TypeScript to integrate and interact with Accor services (VPC, Resto, Sorties, etc.) on web and mobile platforms.. . TAGS: NodeJS - GraphQl - Express Js - TypeScript - React - Docker - Docker Compose - PostgreSQL - AWS(Dynamo, CloudWatch, ECR, ECS, Lambda ...), - Terraform - Splunk - GitLab CI / CD - Jest.

Development and redesign of several microservices that recover audiovisual data from various sources on the Radio France antenna (broadcast, chronicle, musical title, concerts...) and qualify this data to make it available for the other tech teams (front, distribution, back office) for enrichment, utilizing RabbitMQ with Topic Exchange for communication between different microservices. Emission Capture Service: Utilizes FFmpeg to record and encode radio broadcasts into audio streams, then transmits the encoded audio files via RabbitMQ for further processing by other microservices (distribution, metadata enrichment). Metadata Service: Collects details of radio broadcasts (title, description, participants, date, and time of broadcast) via external APIs, stores the structured metadata in a MongoDB database, and uses RabbitMQ to send messages to other microservices for integration or further enrichment. Transformation and Enrichment Service: The back office in Next.js with Redux for state management, integrating reusable components and custom hooks for interactive manipulation of metadata (tags, categories, etc.), ensuring real-time data updates and a smooth user interface for content administration and enrichment. Distribution Service: Uses AWS S3 for media storage, OAuth for authentication with external platforms, and RabbitMQ for asynchronous communication between microservices, enabling the automatic publication of content enriched with metadata on multiple digital channels (YouTube, voice assistants, and RSS feeds). . . TAGS: React - Redux / Redux-Thunk - NodeJS - Express Js - SvelteJS - TypeScript - ES6 - Kubernetes - Kubectl - Docker - Docker Compose - RabbitMQ - PostgreSQL - Prometheus - Grafana - ELK - GitLab CI / CD - Jaeger - Jasmine.

Evolution and maintenance of the "Front TV" audiovisual flow management platform (formerly myCANAL), which includes web applications for administration, channel flow management on receivers and televisions, and viewing statistics and generating reports. - Wrapped the streaming flow received by the Canal+ box with a ReactJS presentation layer in the form of a WebView on the television, using Mickey.js to handle remote control events. This includes decoding and transcoding as necessary with FFmpeg, adding metadata, and enabling smooth manipulation of the stream via WebSocket for real-time communication. -Redesign and optimization of the channel filter system included requirements analysis, implementation of filtering algorithms (by type, date, category, and tag) using Node.js, performance optimization via browser caching to reduce computational load, and enhancement of the user interface with React and Material-UI to ensure a responsive and efficient user experience - Developed the "Program Recommendation" feature, leveraging the user's viewing history exported to a REST API by the data team, to dynamically calculate and present recommended programs via an interface optimized with React.js and Redux. - Tracked TV navigation on channels and programs using JavaScript scripts to capture viewing events, securely sending them in real-time to a collection service like AWS Kinesis Firehose, followed by structured storage of this data in AWS S3. . . . TAGS : React.Js( Redux, React Router, Context API , Flow, React storybook ... ), Nodejs, ExpressJs, GraphQL, Apollo, Mongoose ODM, MongoDB, REST API, JSON Web Token, AJAX, Webpack, ES6, Unit Testing, Mocha, Chai, Jasmine.

Extremely interesting experience with 5E. I have participated in the development of Real-time notifications for an e-learning platform as a task. TAGS: Django, Python, Celery, Django REST framework, JSON Web Token, Django Channels, WebSocket, Redis, RabbitMQ, Vue.js ( to test notifications in the browser ).

Worked as Part-time MEAN.JS Developer to build the lead e-learning plateform in tunisia. TAGS : AngularJs, Nodejs, ExpressJs, Sequelize ORM, MySQL, REST API, AJAX, JIRA, Bitbucket, Test-driven development( Mocha and Chai ).