Senior MLOps Engineer

Bumble is looking for MLOps engineers to join our team and play a key role fulfilling our mission to create a world where all relationships are healthy and equitable. Concretely, this means maintaining and improving the MLOps platform that supports the lifecycle of state of the art machine learning models, developed by several teams at Bumble Inc (Recommendations, Trust & Safety, etc).

With millions of images and messages exchanged on our platform every day, there is a wealth of opportunity to make a real difference in this role and help people to find love all over the world! The ideal candidate combines strong technical skills, extensive experience in managing Kubernetes clusters along with a passion for ML.

WHAT YOU WILL BE DOING

• Maintain and improve the MLOps platform allowing us to serve predictions at massive scale and iterate faster on all our models

• Administer and manage the GPU-powered MLOps Kubernetes clusters

• Be part of the on call rota to support smooth operation of the MLOps platform and the health of all our ML services

• Improve the MLOps platform in terms of processes, performance and testing

• Research and experiment with the latest MLOps technologies and inference frameworks to unlock capabilities for all ML engineers in the company

• Support the efforts of ML engineering and product teams

• Mentor and coach team members on DevOps and ML engineering best practices

WE’D LOVE TO MEET SOMEONE WITH

• Deep understanding of Kubernetes infrastructure and experience administering GPU enabled Kubernetes clusters at scale
• Experience working with Docker and containerised applications
• Experience with at least one programming language such as Python, Golang etc.
• Experience with CI/CD tooling such as ArgoCD, GitHub Actions etc.
• Experience configuring and maintaining monitoring systems such as Grafana, Prometheus etc.
• Experience with IaC tooling, e.g. Terraform
• Comfortable in reacting to incidents and be part of an on call rotation
• Good understanding of machine learning model development life cycle processes and tools: ML model development and experimentation, training pipelines, model serving and monitoring
• Ability to work collaboratively and proactively in a fast-paced environment alongside engineers, scientists and non-technical stakeholders
• A passion for keeping up with the latest ongoings in DevOps and MLOps communities
• A curious mind, self-starter and endlessly keen to learn and develop themselves professionally

AN ADDED BONUS IF YOU HAVE

• Experience configuring and maintaining tools supporting the ML model lifecycle (e.g. Kubeflow)
• Experience configuring and maintaining feature and embedding storage systems
• Experience with cloud infrastructures - GCP is a plus