As a MLOps Engineer, you will work at the intersection of cloud infrastructure, machine learning operations, and software engineering. This is an hybrid role, where you will maintain and evolve our cloud-based infrastructure while building tools, pipelines, and deployment workflows that enable machine learning researchers and engineers to move faster and deploy models reliably into production.
From web apps & microservices running in kubernetes to specialized software communicating with industrial cameras running in constrained factory environments, Scortex software requires a solid and flexible infrastructure.
It is comprised of automated code testing, integration, and deployment pipelines, compute facilities and large-scale storage systems in data centers (cloud), monitoring systems, replicated and sharded databases, on-premise analysis systems deployed in factories, and the networks (VPNs, 4G, industrial protocols etc.) to tie them all.
As a MLOps Engineer, it is expected that you place the utmost importance in automation, monitoring, security, and documentation. In addition, you will collaborate closely with ML researchers, software engineers, and product teams to improve reproducibility, scalability, and operational reliability across the ML lifecycle.
As such, your tasks will include :
Improve & Automate the Provisioning and Scaling of –physical and virtual– infrastructure ( incl. containers, software versions and configurations, databases, cloud resources, networking, user accounts, etc. )
Improve & Automate the Monitoring of our systems with dashboards, collectors and integrations
Optimizations ( of Billed Cost, Dev Time or Latency/Throughput ), ranging from database indexes and configurations to appropriate device selection
Build internal tooling to support data preparation, experiment tracking, model training, and evaluation
Design and maintain scalable model deployment pipelines for production inference systems
Support deployment and monitoring of computer vision models in production environments
Communicate with our team members, document your work and write guides to using the architecture
Stay current, keep up to date with security and systems management tools and practices
You may occasionally visit factory sites, mainly but not limited to France and Europe.
