Site Reliability Engineer / DevOps Engineer
Morpheo is a consortium made of Rythm, école Polytechnique and Université Paris Descartes devoted to developing a secure infrastructure for orchestrating Machine Learning on sensitive data. All our code is open source and we use modern tools such as Docker, Kubernetes and Blockchain technology (HyperLedger). Our computer science team adopted the DevOps philosophy: everybody is involved in both coding and deploying/monitoring code in production.
You will be hired by Rythm a neurotechnology startup which works on sleep analysis and enhancement. Rythm produces the Dreem headband which collects physiological activity during sleep, such as brain activity. We analyse sleep data from multiple sources and build artificial intelligence algorithms to help diagnose various pathologies.
You will be the main system administrator in the Morpheo team.
This mainly involves maintaining a large Kubernetes cluster embedding Blockchain components and running a massive GPU-based machine learning cluster in the cloud.
You may also be involved the development of some components of the platform (Golang, Python & Rust).
One of you key roles will be to spread the DevOps culture among the computer science team (train developers on Kubernetes and Terraform, help them monitor their microservices efficiently and teach machine learners and researchers how to build optimal Docker images).
Skills required (decreasing order of importance)
- Docker & Kubernetes
- AWS / GCP (GKE)
- Flawless knowledge of Git
- Linux system administration
- Experience with continous integration platforms
- Development in Golang
- Blockchain development: Hyperledger Fabric
- Machine learning basics
- Experience with CoreOS is plus