Vu H.

– Led the migration from AWS VM-based architecture to Kubernetes-based stack on GCP. Also contributed to rewriting the bidder platform from TypeScript to Golang to enhance performance. – Designing, provisioning, and maintaining GCP infrastructure from the ground up – Deploying and managing GKE clusters and workloads – Implementing CI pipelines and automation workflows – Establishing best practices for Terraform, Helm, and Golang development – Setting up Datadog monitoring with a focus on observability best practices

– Development of a next-generation core catalog covering all OVH Cloud products and services. Along with several tool chains to migrate massive data from the old platform to the new one. – API and system catalog development – Data model design – High throughput data transfer pipeline development

– Development of a high performance, real-time computed ad-bidder platform which uses public user data for automatic advertising auctions. – API and core domain development – Optimizing bidding algorithm – Distributed system, microservices architecture design – Managing cloud infrastructure with IaC

– Leading the development of a MongoDB-as-a-service on a multi-cloud platform. This project was intended to replace MongoDB Atlas for internal uses to reduce the cost by up to 50%. – The project was made from scratch with an international team. After a year of development, the MVP was delivered on time and we managed to host the first Ubisoft application on our platform. The project was approved for a bigger budget and the team grew double with the goal of hosting major triple-A games in near future. – Developing APIs and Kubernetes operators to provision MongoDB clusters – Helping architect and SRE teams on architecture design – Establishing CI/CD pipelines – Defining milestones, tasks as well as a way-of-working – Discussing with product team, SRE team and architect team to define product scopes – Training new comers on Golang and Kubernetes

– Maintenance and development of a Java application that manages over 100 million inter-banks option contracts. It allows back offices to validate, record and search contracts efficiently. – By the end of the contract, my work on optimising the search engine helped reduce the search time by at least 50%. In some critical cases, search time dropped from 10 minutes to under 1 minute. – Maintenance and features development – Database maintenance – Optimising search algorithms and database queries