Key Notes
Python and SQL knowledge and experience building data pipelines for machine-learning workloads are required. You’ll work within Trainline’s ML team on data models, feature stores and AWS pipelines, collaborating with ML engineers and data scientists. The London role pays £60,000–£67,500, with office attendance for at least 60% of working time over 12 weeks.
What You'll Work On
- Build scalable data pipelines, models and feature stores for Trainline’s machine-learning workloads, working with ML engineers and data scientists.
- Deploy AWS data applications through automated build, test and release pipelines, maintaining observability and performance of production data feeds.
- Work with the wider data engineering and platform community to improve dataset reliability and share approaches to machine-learning data workloads.
Why This Role Matters
- Trainline’s ML engineers and data scientists depend on usable datasets, with this role’s pipelines and feature stores supplying inputs for model training and downstream applications.
- Business dashboards and real-time data products use high-volume event data, making pipeline observability and deployment controls relevant to the reliability of those data feeds.
What They Are Looking For
- Technical: Applicants need working knowledge of Python and SQL plus experience building pipelines for feature engineering and model training.
- Technical: Applicants need data modelling and cloud warehouse or mart experience.
- Experience: Applicants need spark, Airflow or similar orchestration experience in AWS, including batch and real-time patterns.
- Experience: Applicants need helpful experience with Terraform, Docker, CI/CD, Parquet, Iceberg or distributed training frameworks.