Description
Join Graphcore's dynamic Software Infrastructure team as an Infrastructure and MLOps Engineer and play a pivotal role in scaling and managing the company's infrastructure.
You will develop essential tools and services that empower the broader software team, enhancing the build, test, deployment, and productisation processes of Machine Learning Software components.
The Software Infrastructure team provides critical platforms and services for software development teams across the business, including managing the CI platform and services, build engineering, component integration, and packaging and release systems.
Responsibilities and duties:
- Develop, own, and maintain tools and services to support AI research and engineering teams
- Deploy and maintain services with Kubernetes and Docker
- Manage cloud infrastructure using tools such as Terraform
Candidate profile: Essential:
- Knowledge of Python
- Familiarity with cloud services (e.g., AWS)
- Experience managing or developing in Linux environments
- Understanding of CI/CD principles
- Experience using Kubernetes (k8s)
- Experience of one of the following: maintaining machine learning applications, deploying ML orchestration tools (e.g., NV Ray, KFP, SkyPilot), or managing ML accelerator hardware (e.g., DCGM)
Desirable:
- Experience with Infrastructure as Code (IaC) tools (e.g., Terraform/OpenTofu)
- Experience with GitHub Actions
- Experience with modern observability tooling (e.g., Prometheus)
- Experience with Grafana
- Knowledge of Go/Java/C++ (or similar language)