Yassine TIMOUMI

▪ Led the migration of a microservice from Gin to Goa, including redesigning the service architecture, refactoring existing code, and adapting it to the new framework. ▪ Successfully managed data migration and integration with another microservice to ensure system consistency and performance. ▪ Worked on a frontend React.js redesign project, focusing on refactoring components, improving UI/UX, and modernizing the application’s overall look and feel. Technologies: Go, Gin, Goa design, React.js, MySQL, Redis, GCP.

▪ Developed and managed cloud infrastructure, including creating virtual machines, reserving IPs, provisioning volumes, setting up PostgreSQL clusters, and configuring databases. Implemented SSH configuration and role-based access for secure remote connections. ▪ Resolved incidents and debugged issues in the codebase to enhance stability and functionality in a Django-based application. ▪ Implemented user access roles by enhancing the frontend with Nuxt (Vue.js) and integrating role-based access controls in the backend using Python Django. ▪ Enhanced user access rights in the frontend and improved the login and logout components for a smoother user experience. ▪ Implemented authentication functionality using SGIAM to ensure secure access and compliance with organizational standards. Technologies: Python, Django, PostgreSQL, Nuxt (Vue.js), SGIAM, SG Cloud.

My tech working stack: Golang, Grpc, Gin, Goa design, Swagin, Gorm, Python, Rust, Redis, Jwt, OpenAPI v3, GCP, Docker, MySQL Tasks: Ekyo-Rest: add Rest APIs using Goa-desing. Keywords: Go, Goa-design, Mysql, Context, Redis, OpenApi v3. Ekyo-Grpc: create grpc server to send data into Analysis micro service. Keywords: Go, Goa-design, GoaGrpc, Interceptors, OpenApi v3. Analysis-Rest: building from scratch Rest APIs using golang and gin. Keywords: Go, Gin, Swagin, Mysql, Context, OpenApi v3, jwt. Analysis-Grpc: create two grpc clients to receive data from Ekyo-Grpc and Authenticator-Grpc microservices. Keywords: Go, Grpc, Proto 3, Interceptors, OpenApi v3. Tests: create Unit and Integration tests. Keywords: Go, Mock, Testify.

• Webinar-API: creation and building GraphQL APIs using Golang to distribute virtual meetings. Keywords: Go, Cryptography, GraphQl, Mysql, Docker, docker-compose. Link; https://webinarplease.com/fr • Webtv-API: add entities and improve StatBundle & VideoBundle modules. • DevOps : Deployment of projects on k8s. Keywords: Docker, docker-compose, kubernetes, k8s, gitlab-ci. • STATS: Improve the stats module with ELK by changing queries on Kibana. • DAM: create the database architecture and the APIs for the new BO in Go. Keywords: Go, Gin, Gorm, Mysql, GraphQl, Docker, docker-compose. • Video Background AI : Blur and change the background of a video in real time with Python, MediaPipe and React and put the app into production using gitlab ci/cd. Keywords: Python, Tensorflow, React, Typescript, Docker, docker-compose. • Containerization: Dockerization of php, go, react, vue and next projects. • Hecate: Generate beautiful thumbnails, GIFs and summaries from videos. Keywords: C++, OpenCv, Python, Docker, docker-compose. • Denoiser : Real-time speech enhancement. Keywords: Javascript, RNN, Wasm, Jitsi model

Head of Division Computer Vision: - Management of engineering teams - Strategy development and project planning

Smart video surveillance system based on computer vision and deep learning: Our system includes 5 CV applications : -Object detection and tracking (with OpenCV) -Automatic attendance record with facial recognition(Validation accuracy = 99% with face_recognition) -Abnormal activities recognition (Validation accuracy = 94% with ResNet 3D) -Fire and Flood detection (Accuracy = 84% with ResNet) -Emotions detection ( Accuracy on the dataset of test = 66% with our CNN architecture) #videosurveillance #computervision #artificialintelligence #machinelearning #deeplearning #webdevelopment #api #flask #reactjs #dvr #nvr #ipcamera #sms #twilio #jwt #security #opencv #resnet #cnn #hog #frontend #backend #postgresql #pycharm

Project : Semantic segmentation for teeth.  Keywords: Deep Learning, Python, Pytorch, Pycharm, CUDA C++. Tasks carried out: Data cleaning, data engineering, training, and testing the model. Algorithmes: pointnet2/pointnet++, farthest point sampling, plydata to hdf5 file. Data: HDF5, PLY.

Smart Grid revolution