You will be responsible for building and maintaining the models and data pipelines behind our data science workflows, ensuring the accuracy, consistency, and efficiency of the data used for training and inference, working across structured and unstructured data from many sources on a large-scale, distributed platform.
Responsibilities:
Lead, mentor, and grow a team of MLOps engineers, owning delivery and technical quality.
Take end-to-end ownership of infrastructure and pipeline initiatives across the LMM group, from design through production.
Stay hands-on: contribute to design and code, review work, and set engineering standards.
Drive the team through critical milestones in ML model-lifecycle and infrastructure ownership.
Partner with R&D and other stakeholders to translate research needs into robust, scalable systems.
Help evolve the platform, including our ongoing migration from Dask to Ray.
BSc or Master's degree in Computer Science, Mathematics, or Engineering.
At least 5 years of commercial experience in Python.
At least 3 years of hands-on commercial MLOps experience in production (not side projects).
Experience managing or leading a team of engineers, with ownership of both people and delivery.
Hands-on experience owning the ML model lifecycle (training, deployment, monitoring, retraining).
Experience with pipeline orchestrators such as Dagster or Airflow.
Experience with a major cloud provider such as GCP, AWS, or Azure.
Experience with distributed computing systems.
Experience with Docker.
Experience with Kubernetes.
Commercial experience writing and maintaining scalable ML systems.
Fluent in English, both written and spoken.
